English. Instructions for Use. NM Application Suite. Extended Brilliance Workspace NM 1.0 PHI

Transcription

1 English Instructions for Use NM Application Suite Extended Brilliance Workspace NM 1.0 PHI

2 NM Application Suite INSTRUCTIONS FOR USE Release 1.0 English Philips Healthcare Rev A PAI

3 Warranty and Copyright Statement Philips Healthcare has taken care to ensure the accuracy of this document. However, Philips Healthcare assumes no liability for errors or omissions and reserves the right to make changes without further notice to any products herein to improve reliability, function, or design. Philips Healthcare provides this guide without warranty of any kind, either implied or expressed, including, but not limited to, the implied warranties of merchantability and fitness for a particular purpose. Philips Healthcare may make improvements or changes in the product(s) and/or program(s) described in this manual at any time. This document contains proprietary information which is protected by copyright. All rights are reserved. No part of this publication may be reproduced, transmitted, transcribed, stored in a retrieval system or translated into any human or computer language in any form by any means without the consent of the copyright holder. Unauthorized copying of this publication may not only infringe copyright but may also reduce the ability of Philips Healthcare to provide accurate and up-to-date information to users and operators alike. Philips Healthcare reserves the right to revise this publication and to make changes in content from time to time without obligation on the part of Philips Healthcare to provide notification of such revision or change. Trademarks ACCESS TM, ADAC, ALLEGRO, ARGUS, Atlas TM, AutoQUANT, AutoSPECT, AutoSPECT Plus, CardiaQ, CardioMD, Cardio TM, Cardio TM 60, CardioTrac TM, CCT TM, ColliMATE TM, CPET, ENsphere, EPIC TM, ExSPECT TM, EZX TM, FlexLOGIC TM, Forte TM, GEMINI TM, GENESYS, GlobalQ, InStill, InteLOGIC TM, JETSphere TM, JETStream, MCD/AC TM, Midas TM, MOSAIC TM, P3IMRT, P 3 MD, Pegasys TM, PINNACLE3, PIXELAR, Precedence TM, SENTRY TM, Shadow TM, SKYLight, SKYTable TM, SMARTSIM, Solus TM, TeleLOGIC TM, Transcam TM, Vantage TM, VersaTable TM, Vertex TM, WebView TM, and X-ACT TM are trademarks or registered trademarks of Philips Healthcare. Adobe, the Adobe logo, Acrobat, the Acrobat logo, and PostScript are trademarks of Adobe Systems Incorporated or its subsidiaries and may be registered in certain jurisdictions. Sun, SunView, NFS, OpenWindows, Solaris, and SPARCstation are trademarks of Sun Microsystems, Inc. SPARC is a registered trademark of SPARC International, Inc. SPARCstation is a trademark of SPARC International, Inc. licensed exclusively to Sun Microsystems, Inc. UNIX and OPEN LOOK are registered trademarks of UNIX System Laboratories, Inc. X Window System is a trademark of the Massachusetts Institute of Technology. Microsoft and Windows are either registered trademarks or trademarks of Microsoft Corporation in the United States and/or other countries. Symantec and Symantec pcanywhere are U.S. registered trademarks of Symantec Corporation. Symantec AntiVirus and Symantec WinFax PRO are trademarks of Symantec Corporation. Other brand or product names are trademarks or registered trademarks of their respective holders. Prescription Device Statement Caution: Federal law restricts this device to sale by or on the order of a physician (or properly licensed practitioner). Disclaimer Neither Philips Healthcare, its parent, nor any of its worldwide affiliates shall be liable or obligated in any manner in respect of bodily injury and/or property damage from the use of the system/software if such is not in strict compliance with instructions and safety precautions contained in the relevant operating manuals and in all supplements thereto, in all product labels, and according to all terms of warranty and sale of the system, or if any change not authorized by Philips Healthcare is made to the software operating the system. CE Marking NM Application Suite is CE Marked to the Medical Device Directive 93/42/EEC. Manufacturer: Philips Medical Systems (Cleveland), Inc North First Street San Jose, California European Authorized Representative: Philips Medical Systems Nederland B.V. PMS Quality & Regulatory Affairs Veenpluis PC Best The Netherlands Philips Healthcare NM Application Suite Instructions for Use Document number Rev A 2009 Koninklijke Philips Electronics N.V., 3860 North First Street, San Jose, CA 95134, USA Printed in the United States of America

4 Contents 1 Introduction Intended Use Conventions Used in This Manual Overview Intended Users... 2 Philips Healthcare Rev A 2 General Information Getting Online Help Getting Remote Help from Philips Loading Files Using the Control Panel Using the Patient Selector Using the Workflow Using the Data Managers Using the Global Image Tools Drawing ROIs Editing ROIs Reusing ROIs Changing the Edge Detection Type Using Isocontours Drawing Other ROI Shapes Other ROI-related Tasks Creating a Custom ROI Using Viewers Viewer Components Linking Viewers Moving and Resizing Viewers Using a Curve Viewer Using Layouts Editing Layouts Editing Layouts in AutoSPECT Pro AutoSPECT Pro The AC Map Workstep Reviewing an AC Map Using CT-AC Data Using Vantage AC Data Using Chang s AC The Reconstruction Workstep Start/End Zoom NM Application Suite Release 1.0 iii

5 3.2.3 Method Bound Iterations Matrix Size Subsets Start Attenuation Correction Scatter Correction Decay Correction Motion Correction Axis Correction Filter Y Axis Cutoff Order Comparing Parameters Using Manual Motion Correction Analyzing Sinograms and Cyclograms The Reorientation Workstep Reorienting Datasets Setting Preferences for AutoSPECT Pro General Preferences Preferences for the Setup Workstep Preferences for the ACMap Workstep Preferences for the Reconstruction Workstep Preferences for the Reorientation Workstep Preferences for the Review Workstep Saving and Applying Preferences AutoSPECT Pro Default Protocols Cardiac MUGA Using MUGA Results Preferences First Pass Using First Pass Results Preferences Shunt Using Shunt Results Preferences Review Layouts Philips Healthcare iv NM Application Suite Release 1.0

6 5 Whole Body Using Three Phase Analysis Results for Three Phase Analysis Parameters for Three Phase Analysis Using Ileosacrum Ratio Computation Results for Ileosacrum Ratio Computation Review Layouts Lung Using Lung Washout Results Ventilation Results Perfusion Results Review Layouts Philips Healthcare Rev A 7 Renal Using ROIs in Renal Supplying Inputs for Renal Renal Results Preferences Review Layouts Simple Renogram Using Simple Renogram Results Pre Post Lasix Using Pre Post Lasix Results Post Renogram Lasix Using Post Renogram Lasix Results GFR Gates Using GFR Gates Results ERPF Schlegel Using ERPF Schlegel Results ERPF Schlegel with Void Using ERPF Schlegel with Void Results ERPF MAG Using ERPF MAG Results NM Application Suite Release 1.0 v

7 7.13 Deconvolution Using Renal Deconvolution Results DMSA Static Ratio Results Hilson Index Using Hilson Index Results Patlak Using Patlak Results Cortical Analysis Using Cortical Analysis Results Split T 1/ Using Split T1/ Results Endocrine Thyroid Using Thyroid Results Parathyroid Using Parathyroid Subtraction Results Review Layouts Esophagus Using Esophagus Esophagus Using Esophagus Esophagus Results Preferences Gastro-Esophagus Reflux Using Gastro-Esophagus Reflux Gastro-Esophagus Reflux Results Preferences Review Layouts Gastro Intestinal Using Gastric Emptying Results Preferences Review Layouts Philips Healthcare vi NM Application Suite Release 1.0

8 11 Hepatobiliary Using GallBladder Results for GallBladder Preferences for GallBladder Results for GBEF Static Analysis Preferences for GBEF Static Analysis Review Layouts General Review QC Tools Using QC Tools UFOV and CFOV Results Tabs Using the QC Tools Toolbar Saving Results Saving to a Text File Saving as Secondary Capture Saving to Film Astonish Reconstruction When to Use Astonish Using Astonish Reconstruction Filtering and Noise Iterations and Subsets Applying Other Corrections Further Reading Information on OSEM reconstruction Information on resolution recovery Philips Healthcare Rev A 15 Backing Up and Restoring Data Backing Up Data Restoring Data NM Application Suite Release 1.0 vii

9 viii NM Application Suite Release 1.0 Philips Healthcare

10 1 Introduction The NM Application Suite is processing software designed to streamline Nuclear Medicine workflow to meet productivity demands of NM imaging departments. It includes a comprehensive suite of planar, SPECT and QA applications and provides users with the ability to make a play list of protocols on the fly and access the tools when needed, where needed. 1.1 Intended Use The NM Application Suite is a nuclear medicine image display and processing application suite that provides software applications used to process, analyze, and display medical images and data. The results obtained may be used as a tool, by a nuclear physician, in determining the diagnosis of patient disease conditions in various organs, tissues, and other anatomical structures. The data processed may be derived from any nuclear medicine gamma camera. The NM Application Suite should only be operated by qualified healthcare professionals trained in the use of nuclear medicine equipment. 1.2 Conventions Used in This Manual This manual uses the following conventions: Philips Healthcare Rev A WARNING: Messages that alert you to conditions that may result in death or serious injury. CAUTION: Messages that alert you to conditions that may result in one or more of the following: Minor or moderate injury to you or the patient Damage to the equipment or other property Data loss IMPORTANT: Vital information that describes how to properly install, configure, or use the system. Note: Additional information that may help explain an action or procedure The following items appear in a unique style in the text: <Enter> Press the Enter key on the keyboard. <Return> Press the Return key on the keyboard. NM Application Suite Release 1.0 Introduction 1

11 1.3 Overview Screen elements and system controls (menu items, buttons, switches, etc.). For example, Save. Computer messages displayed on the screen. For example, Reboot After Installation. All host names, IP addresses, and Ethernet addresses displayed in these procedures are examples only. Before entering host names, IP addresses, or Ethernet addresses, check the existing network configuration to make sure that you do not duplicate any existing entries. 1.3 Overview Intended Users The NM Application Suite is a Windows-based nuclear medicine processing workstation that functions seamlessly out of the box with JETStream acquisition. The product's design and features improve operator and department workflow by allowing workflow distribution of technologist and physician activities to workspots distributed within nuclear medicine departments in hospitals, cardiology offices, imaging centers and remote locations. This includes the application suite running on the Extended Brilliance Workspace system. The NM Application Suite can also communicate with competitor cameras and workstations and interfaces seamlessly with multi-modality data and PACS. The NM Application Suite offers a complete, robust nuclear medicine workstation that meets the needs of customers in all geographies. Extended Brilliance Workspace (EBW) is the Philips CT and PET/CT processing workstation solution. EBW is Windows based and uses high performance hardware for handling large CT datasets. The Medical Imaging Platform (MIP) is a set of standardised Philips-wide Windows-based software assets that are being used to form the basis of next generation imaging applications. The NM Application Suite uses the EBW as a platform to integrate SPECT, PET and CT applications onto a single desktop for our Hybrid Imaging customers (using Precedence and BrightView XCT). In addition, the NM Application Suite leverages several of the resources provided by EBW such as patient management, printing, archival networking, and an application bridge to maximize throughput and quality. NOTE: If you load third-party software onto the workstation it may decrease system performance. Philips Healthcare NM Technologists will perform acquisition and processing of image data generated from a gamma camera. Their experience can range from beginning nuclear medicine student to multiple years of experience. The system must therefore be simple to use and understand for the 2 Introduction NM Application Suite Release 1.0

12 Overview 1.3 inexperienced user, yet flexible enough to accommodate the experienced user. Use of the system by inexperienced users may or may not be under the supervision of fully-trained personnel. Physicians will perform processing, post-processing, display/review and interpretation of nuclear medicine images acquired from a gamma camera. Experience can range from 1st year medical student, resident to board certified Nuclear Medicine physician. The system must therefore be simple to use and understand for the inexperienced user, yet flexible enough to accommodate the experienced user. Use of the system by inexperienced users may or may not be under the supervision of fully-trained personnel. Physicists will perform processing, post processing, display/review for acceptance testing and QC of Nuclear Medicine images acquired from a gamma camera. These users require analysis tools, such as manual regions of interest, statistics inspection, and curve generation beyond the standard tools found in clinical applications. Philips Healthcare Rev A NM Application Suite Release 1.0 Introduction 3

13 1.3 Overview Philips Healthcare 4 Introduction NM Application Suite Release 1.0

14 2 General Information This section describes features that are available for all applications. The layout of the NM Application Suite consists, broadly speaking, of two areas: Figure 1 NM Application Suite layout Philips Healthcare Rev A 1. Control Panel 2. Viewing area The Control Panel contains all the tools necessary to use the applications. See the section on Using the Control Panel on page 8 for details. The viewing area consists of viewers that contain study images, results, curves, and any other visual information presented by an application. See the section on Using Viewers on page 63 for details about viewers. It also has a strip at the top that contains DICOM information from the study. If you right-click in the viewing area, a menu appears that provides access to many of the tools available in the Control Panel. In the DICOM strip, you can select the information to display by right-clicking and selecting a field from the pop-up menu. Some of these may be hidden when you use the Deidentify feature in the Utilities manager. NM Application Suite Release 1.0 General Information 5

15 2.1 Getting Online Help 2.1 Getting Online Help As with other EBW applications, online help is available by clicking the Help button at the top of the EBW window. This displays the Philips splash page with some buttons at the top. Select your application s manual in the drop-down menu under the Operation Manual button, and then click Operation Manual to open the file. 2.2 Getting Remote Help from Philips For some problems, you may need to provide Philips Healthcare with remote access to your system. You can do this using the EBW Remote Service feature: 1. In EBW, click Help. 2. In the next screen, click Remote Service. 3. In the Remote Access dialog, select View Only or Remote Control Sharing, as required, and click Start. This begins the remote session. 4. To end a session, click Disable, near the upper right of the screen. 2.3 Loading Files When you load studies into the NM Application Suite, you can choose to load them through EBW s Review or Analysis menus. Loading a study through the Review menu opens it in the selected application in the Review (last) workstep, allowing you to view all the datasets. However, you can still select the Setup workstep and proceed through all the worksteps to reprocess the study. NOTE: If you go to the Setup workstep, the Review workstep is not available unless you load the buckets and process the study. Loading through the Analysis menu opens it in the selected application s workflow, allowing you to start processing. Philips Healthcare 6 General Information NM Application Suite Release 1.0

16 Loading Files 2.3 Philips Healthcare Rev A Figure 2 Location of EBW Review and Analysis menus In both cases, the process is to select one or more studies in the EBW Patient Directory and then select one of the applications from the Review or Analysis pull-down menus. Both menus contain these NM Application Suite applications: NM Application Suite: This opens the Review application, allowing you to review all the data before making any edits. NM Cardiac: This includes methods for MUGA, First Pass, and Shunt. NM Whole Body: This includes methods for Three Phase Analysis and Ileosacrum Ratio. NM Renal: This includes methods for Renogram, ERPF MAG3, Lasix, Patlak, and others. NM Lung: This includes methods for Washout, Ventilation, and Perfusion. NM Endocrine: This includes methods for Thyroid and Parathyroid. NM Hepatobiliary: This includes methods for Gallbladder (Hepatic Duct, Common Bile Duct, Duodenum, and Background) and GBEF Static Analysis. NM Gastric Emptying: This includes methods for Liquid and Solid. NM Esophagus: This includes methods for Gastro-Esophagus Reflux and Esophagus. AutoSPECT Pro: This includes methods for Cardiac, Vantage, Thallium, Astonish, Brain, Bone, and others. NOTE: You can load multiple patients, but as the number increases, performance may suffer, depending on factors such as file size, number of series, etc. See the NM Application Suite Reference Manual for information on minimum loading requirements for specific applications. NM Application Suite Release 1.0 General Information 7

17 2.4 Using the Control Panel When a study loads, the NM Application Suite tries to match the study information to some matching criteria. If a match is found, the study loads into the Define Regions workstep, allowing you to start drawing ROIs immediately. If it fails, it loads into the Setup step, allowing you to assign data to the correct buckets manually before proceeding to the Define Regions workstep. For information on setting up match criteria, see the section on Editing Auto Matches on page Using the Control Panel The Control Panel is where most of the work in the NM Application Suite takes place. There are four areas of the Control Panel: Philips Healthcare 8 General Information NM Application Suite Release 1.0

18 Using the Control Panel 2.4 Philips Healthcare Rev A Figure 3 Control Panel 1. Patient Selector: If you have loaded multiple patients, you can use this drop-down menu to select one. See the section on Loading Files on page 6 for details. NOTE: When you select a study using the drop-down menu, the workstep information for the current study is preserved. For example, if you are on the Review Results workstep in a study, then select a different study, if you use the pull-down menu to go back to the first study you automatically return to the Review Results workstep. You do not need to reselect data and redraw ROIs. 2. Workflow Navigator: This allows you to navigate among the worksteps, and perform work and choose tools that are specific to a workstep. NM Application Suite Release 1.0 General Information 9

19 2.4 Using the Control Panel Here you can load datasets, draw ROIs, review results, etc. See the section on Using the Workflow on page 10 for details. 3. Data Managers: These allow you to choose activities that are unrelated to a workstep, for example setting preferences, performing image manipulation, selecting an application, etc. See the section on Using the Data Managers on page 16 for details. 4. Global Image Tools: This is where you can choose among tools that are common across applications, like saving, printing, changing the colormap, etc. See the section on Using the Global Image Tools on page 46 for details. You can minimize and maximize the Control Panel using the pushpin icon in the top right of the panel. Figure 4 Control Panel pushpin When it is minimized, you still have access to the workstep buttons. You can reapply the current preference by clicking Restart, at the bottom of the Control Panel. You might do this to reload the study and reset all ROIs. To exit the NM Application Suite, click Exit at the bottom of the Control Panel. Use the View Status Bar button to display the status bar. The status bar appears at the bottom of the viewing area, and contains information pertaining to the current workstep. You can drag the top of the status bar to make it bigger, allowing you to see multiple lines. Figure 5 Status bar Using the Patient Selector The Patient Selector allows you to select a new patient to display. If you loaded multiple patients into the NM Application Suite, you can use the drop-down menu to select a patient to work on. Philips Healthcare Using the Workflow There are three buttons that control the workflow: 10 General Information NM Application Suite Release 1.0

20 Using the Control Panel 2.4 Figure 6 Workflow buttons 1. Next and Previous buttons 2. Workstep list The left and right arrows at the sides (#1 in the image above) go to the next or previous workstep. The small down arrow at the bottom of the workstep name (#2 above) provides a pull-down list of all the worksteps, allowing you to go directly to any available workstep. Use the buttons to navigate through the worksteps. The basic workflow is described immediately below; a more detailed description follows. Load data into an application: Click an application icon in EBW; the default preferences are loaded (but you can reselect another set manually later). Define Regions: If automatching succeeds, draw ROIs; if not, manually load the data buckets, then proceed to Define Regions. NOTE: The workflow for AutoSPECT Pro is slightly different at this point. In that application, instead of defining regions and reviewing results, you perform reconstruction and reorientation (and possibly AC map generation). See the chapter on AutoSPECT Pro for details. Review results: After drawing the ROIs, review the processed results. Philips Healthcare Rev A Reviewing Data CAUTION: Be sure to verify any automatically generated results in an application. You are responsible for ensuring that the results are consistent with your clinical expectations. Review the processed images. When you load data into the NM Application Suite application (as opposed to a dedicated application such as Renal, Cardiac, etc.), it is displayed in the Review application, allowing you to see all the data before loading it into an application. Depending on the study type, various layouts are available in which to view the data. You can use the full functionality of the Control Panel without loading the data into another NM Application Suite Release 1.0 General Information 11

21 2.4 Using the Control Panel application. This means that you can perform image processing functions with the ICMT or apply color maps, for example. For more on the Review application, see the section on General Review on page 157. NOTE: Studies are loaded in the Review application only when you select the NM Application Suite icon in the EBW Patient Browser. If you select another application instead, the study opens in the Define Regions workstep if automatching succeeds, or the Setup workstep if it fails. Setup Workstep (Loading Data) If the data meets the application s default preference s automatching criteria, the datasets are automatically assigned to data buckets, and the application proceeds to the Define Regions workstep. If not, you can load the data manually by using the drop-down list of datasets at the right of the data bucket (#2 in the image below). Figure 7 Data buckets 1. Expected data indicator 2. Assign data to bucket Expected data are indicated by an exclamation point (#1 in the image above). If you do not load the expected data, some operations or results may not be available. NOTE: If you load data manually, be sure to load the data into the correct buckets. If you load the wrong data, the results will be incorrect. There is a Custom Display container at the bottom of the list. This contains 25 buckets. You can use these for data that may not be part of the dataset proper, such as images created with the ICMT, for example. As with the other buckets, datasets can be automatically assigned based on automatching criteria. If no buckets are loaded, the Custom Display layout will not be available in the application. The drop-down list of datasets provides information about what data can be loaded into the current bucket (refer to the figure below): Philips Healthcare 12 General Information NM Application Suite Release 1.0

22 Using the Control Panel 2.4 Figure 8 Loading data 1. An allowed dataset 2. Currently loaded dataset 3. Unavailable dataset 4. Clear Bucket Data that are available to load are not shaded. The currently loaded dataset is lightly shaded. Unavailable data is darkly shaded. To unload a bucket (remove the dataset), click Clear Bucket (#4 above). Define Regions Workstep In this step, you are prompted to draw all ROIs that are required by the current preference. For details on drawing ROIs, see the section on Drawing ROIs on page 59. Review Results Workstep Philips Healthcare Rev A After processing the data based on the ROIs, one or more results layouts are available in the Review Results workstep. They appear in a list under the workstep name. Use the Next button to proceed through them in order. Each application may have unique results. See the chapters on the individual applications for details. Depending on the layout you are using, some viewers may not be included in the results display. To see if there are any hidden viewports, click the Show Hidden Viewers tool in the Utilities manager in the Image Tools. This contains a list of all the viewers that are hidden. If you need to display a hidden viewer, you can either add it to the first viewer in the layout as a new tab, or create a new layout that has a tiled space for it, and save the layout in a Preference. For details on adding tiles to a layout, see the section on Editing Layouts on page 69. For details on creating and editing a layout, see the section on Creating and Editing Preferences on page 43. NM Application Suite Release 1.0 General Information 13

23 2.4 Using the Control Panel Review Workstep The last step allows you to review the processed data in multiple layouts, which depend on the type of study loaded. Editing Auto Matches You can use the Data Matcher to edit matching criteria for a data bucket. To use the Data Matcher, you define expressions to use for various DICOM attributes, and then combine those expressions using logical operators. You can also combine operations. Each expression and each operation is on a separate line in the dialog. NOTE: Unless you are familiar with the syntax and construction of regular expressions, do not attempt to create a new filter from scratch. Just edit existing filters. 1. In the Preferences Data Manager, click the green Filter bucket data button to open the Data Matcher. Figure 9 Filter bucket data button Philips Healthcare 14 General Information NM Application Suite Release 1.0

24 Using the Control Panel 2.4 The Data Matcher window appears: Figure 10 The Data Matcher window 1 Add expression 2 Delete Expression 3 Composite expression 2. In the list of buckets on the left, click the bucket to change. NOTE: Use the Reset button at any time to reset the bucket s expression to its original default values. To cancel, click the Close button. Philips Healthcare Rev A 3. Use the DICOM pull-down menu to select the DICOM attribute to match. 4. Click the + icon (Add expression) near the top right of the window to add the expression to the list. The expression appears in a new line at the bottom of the list. Click a line to display it in the window at the bottom in the form of a search expression. To delete an expression, click the line and click the x icon (Delete Expression) near the top right of the window. 5. Type the string to match in the text box and use the pull-down menu to set the following: Equals - The expression is true if it matches the string exactly. Contains - The expression is true if it contains the string anywhere in it. NM Application Suite Release 1.0 General Information 15

25 2.4 Using the Control Panel Not Equals - The expression is true if it does not match the string exactly. 6. To combine two lines, select the lines using Shift-click to add a line. Then use the Combine pull-down menu to set the operator for the expression: AND - An expression is true if both elements (this one and the one to which it is being added) are true. OR - An expression is true if either element (this one and the one to which it is being added) is true. NOT - An expression is true if this element (this one and the one to which it is being added) is not true. Click the + icon (Add expression) to add the new line. 7. To add more attributes or lines, repeat the process. You can create multiple filters within the window; only one will be applied at any time. 8. Select an expression to use and click Save; the saved expression is highlighted in green. 9. Click Close to close the window Using the Data Managers There are five Data Managers. They organize activities in these categories: Object Browser Object Browser: This allows you to view the available datasets when you are in an application. See the section Object Browser on page 16 for details. Viewer Tools: This contains tools you can use in the image viewers, like pan, zoom, cine controls, etc. See Viewer Tools on page 17 for details. Application Palette: This contains links to all the applications, allowing you to select an application or switch between applications. You can switch at any time, in any application. When you switch, the state of the current application is not saved; it is equivalent to exiting the application and starting a new one. ICMT (Image Curve Manipulation Tool): This provides access to the ICMT. See Image and Curve Manipulation on page 25 for details. Preferences: This allows you to configure, create, and manage preferences. See Creating and Editing Preferences on page 43 for details. The Object Browser is a simple tool that provides a list of the current datasets when you are in an application. However, right-clicking on a dataset allows you to get DICOM information for it, add a viewer to a Philips Healthcare 16 General Information NM Application Suite Release 1.0

26 Using the Control Panel 2.4 layout (in 2D applications), or select a layout to display datasets (in AutoSPECT Pro). You might want to add a viewer, for example, to create a second version of an image to experiment with and compare to the original. The list of layouts in AutoSPECT Pro allows you to control how the data is displayed at the time of selection. To add a viewer in 2D applications: 1. Click a viewer tab to select a container to hold the new viewer. 2. In the Object Browser, right-click on the image to add. 3. Select Add from the menu. The new viewer is added to the container that has the currently selected viewer. To display datasets in a layout in AutoSPECT Pro: 1. In the Object Browser, left-click to select data (using Shift or Ctrl to add to the selection). 2. Right-click on one of the selections and select a layout from the list. NOTE: Only layouts that can accommodate all the selected data are listed. To get DICOM information for a dataset, right-click on it and select Dicom Information from the menu. This displays a window with the most useful DICOM fields and their values. To get the full DICOM information, you must use the EBW Patient Directory or the EBW Subselection feature. Viewer Tools This manager provides tools that affect the viewers. The following is a complete list of the tools available in the manager. NOTE: The tools that are available depend on the viewer that is selected and the type of dataset it contains. Not all tools are available for all viewers and datasets. Tool Description Philips Healthcare Rev A Play/Stop - This starts and stops a cine. Use the entry boxes to set the start and end frames Sync Cine in Viewers - This starts running the cines in all viewers at frame 1, so they all run in sync. Change Frame Rate - This sets the frame rate of a cine in frames per second (fps). Use the slider to increase or decrease. The range is fps. Movie Mode - This determines how a cine behaves. Cyclical Forward repeatedly runs the cine forward. Cyclical Backward repeatedly runs the cine backward. Bounce runs the cine alternately forward and backward. NM Application Suite Release 1.0 General Information 17

27 2.4 Using the Control Panel Tool Description Change bin/time number and Change gated/dynamic slice - For gated tomo datasets these toggle between sliders to select bins or slices. Scroll - For cine datasets, this scrolls through the images in order. Use any of the options in the table below to scroll through a series. NOTE: Deselecting the Scroll menu item only affects the click-drag scroll method, not the arrow keys, scroll wheel, etc. Select Phase - For multiphase studies, use this pull-aside menu to select the phase to display. Astonish - Planar Astonish is an image restoration process (a special type of filter) to enhance the image resolution for a planar image by using the Blind Deconvolution technique. Image restoration is a process to remove or minimize known degradations in an image. There are three settings: Sharp, Sharper, and Sharpest. These correspond to 3, 5, and 10 iterations, respectively. For details on Astonish, see the appendix, Astonish Reconstruction on page 163. NOTE: It is not recommended to use planar Astonish with projection data. Resolution recovery for SPECT data should be done with Astonish in AutoSPECT PRO. Hide/Show ROI - This toggles the display of ROIs. Triangulation - This allows you to use the location of a cursor in one view to alter other views. Triangulation is available for applications that display multiplanar views of the same image set (TRANS/SAG/COR, for example). When you click on an image, a crosshair appears. Dragging the crosshair changes the other views according to where the cursor is. Orientation - This specifies the orientation of the image. Select Visual Type - For volumes, this allows you to select a view of the volume data. Normalize to Frame - This normalizes each individual frame based on its own minimum and maximum pixel value. See the note below. Normalize to Series - This normalizes all frames based on the minimum and maximum pixel value that is found in all frames of a series (the entire volume for a volume dataset or the entire set of frames for a dynamic dataset). NOTE: Use the normalization functions carefully because they can have dramatically different effects, depending on the dataset. When the counts are generally low in some frame in a series, if you normalize the frame using Normalize to each Frame, it (and every other frame) looks like it has a full range of counts. This means that you may not be able to make a useful comparison between it and other frames. Using Normalize to Series preserves the relationship. Philips Healthcare 18 General Information NM Application Suite Release 1.0

28 Using the Control Panel 2.4 Tool Description Define Tiling - This allows you to set the number of (sequential) images in a viewer. The maximum is 4x4 tiles. Previous Page - This displays the previous frame in a cine. Next Page - This displays the next frame in a cine. Here are the options for the Scroll feature above: Scroll forward Click-drag down Right arrow key Page Down key Scroll mouse wheel backward End key (scrolls forward to last frame) Scroll backward Click-drag up Left arrow key Page Up key Scroll mouse wheel forward Home key (scrolls back to first frame) Context Image Tools Right-clicking in a viewport when it is selected provides access to a menu with some combination of the following tools, depending on what type of viewport is active. NOTE: To select a viewport, Ctrl-click it. Tool Description Philips Healthcare Rev A Scroll - For cine datasets, this scrolls through the images in order. Use any of the options in the table below to scroll through a series. NOTE: Deselecting the Scroll menu item only affects the click-drag scroll method, not the arrow keys, scroll wheel, etc. Pan - This moves the image within the viewer. The image follows the motion of the mouse. Zoom- This increases and decreases the apparent size of the image. Drag the mouse up to enlarge the image, down to shrink it. Graylevel - This adjusts the brightness and background, allowing you to normalize a dataset. Hold down the left mouse button and drag it vertically to adjust the brightness in the selected viewer, and horizontally to adjust the background. See the section on Using the Image Control Bar (ICB) on page 47 for details. NM Application Suite Release 1.0 General Information 19

29 2.4 Using the Control Panel Tool Description Position - This allows you to move through the slices of a 3D volume. Roll - This allows you to rotate a 3D image around the axis that is perpendicular to the cursor direction. Rotate - This allows you to rotate a 3D image around the axis that points into the screen. Bin Selection - For gated tomo datasets, you can use this pull-aside menu to select the bin to display. Previous Page - This displays the previous frame in a cine. Next Page - This displays the next frame in a cine. Define Tiling - This allows you to set the number of (sequential) images in a viewer. The maximum is 4x4 tiles. Colormap Tools - This pull-aside menu allows you to select a colormap from a list, an intensity correction from a list, and the next and previous colormap and intensity. [no icon] [no icon] Previous Intensity - This applies the previous item in the intensity list. Next Intensity - This applies the next intensity map in the list. Invert Graylevel - This inverts the pixel values on the underlying image, so darker areas become lighter, and lighter areas become darker. Depending on the colormap applied to the image, you may see a different effect. NOTE: This is not available in AutoSPECT Pro. Histogram Windowing - This resets the viewer contrast level to a reasonable value for data that does not have SUV values calculated. Use this tool to make a quick contrast adjustment when the values are far off normal. Contrast Stretch - This resets the grayscale values of only the pixels in the viewport such that the lowest pixel value becomes 0 and the highest becomes 255. The rest of the pixels are scaled appropriately. This is most useful when you are zoomed into an image far enough to have some pixels beyond the edges of the viewport. Philips Healthcare 20 General Information NM Application Suite Release 1.0

30 Using the Control Panel 2.4 Tool Description Text Box - This allows you to type text on the image. To use it, click in the viewer and type. You can then move it by dragging, and set its properties (see Applying Properties Changes to an Object on page 59 for details). Right-clicking on it displays a menu of options you can use for the object. You can append the text, and by deleting text from the end backwards, you can replace text. This text is not saved; to preserve it you must create a secondary capture. To create text you can save, you must create a Viewer Label (see below). This does not work for 3D images in the MPR Fusion layout. CAUTION: When you add annotations, be careful not to accidentally cover some patient information. If you do, you could misidentify a patient, potentially causing misdiagnosis. Arrow + TextBox - This is like the Text Box above, but it creates an arrow with text attached to it. To use it, click in the viewer where the arrowhead should be, then click to set the other end of the arrow. Now you can type the text. You can move it by dragging, and set its properties. Right-clicking on it displays a menu of options you can use for the object. This does not work for 3D images in the MPR Fusion layout. CAUTION: When you add annotations, be careful not to accidentally cover some patient information. If you do, you could misidentify a patient, potentially causing misdiagnosis. Measure - This pull-aside menu provides tools for making measurements on an image. These tools are also available in the Utilities section of the Global Image Tools. See the section on Making Measurements on page 52 for details. NOTE: This does not work for 3D images in the MPR Fusion layout. Astonish - Planar Astonish is an image restoration process (a special type of filter) to enhance the image resolution for a planar image by using the Blind Deconvolution technique. Image restoration is a process to remove or minimize known degradations in an image. There are three settings you can use: Sharp, Sharper, and Sharpest. These correspond to 3, 5, and 10 iterations, respectively. For details on Astonish, see the appendix, Astonish Reconstruction on page 163. NOTE: It is not recommended to use planar Astonish with projection data. Resolution recovery for SPECT data should be done with Astonish in AutoSPECT PRO. Movie Control - This toggles the display of the movie controls, which allow you to stop and start a cine, and to set the frame rate. Philips Healthcare Rev A Orientation Markers - This toggles the display of single-letter orientation labels appropriate to the viewer type (Left-Right, Anterior-Posterior, etc.). Paste - This pastes a saved ROI created using a Measurement tool. Viewer Label - This displays the Viewer Label Editor. See the section on Using the Viewer Label Editor below for details. Normalize to Frame - This normalizes to the maximum pixel value in the superior half of the pixel data space in each frame. See the note below. NM Application Suite Release 1.0 General Information 21

31 2.4 Using the Control Panel Tool Description Normalize to Series - This normalizes the image using the hottest pixel in the entire slice volume (for volume datasets) or series (for dynamic datasets). NOTE: Use the normalization functions carefully because they can have dramatically different effects, depending on the dataset. When the counts are generally low in some frame in a series, if you normalize the frame using Normalize to each Frame, it (and every other frame) looks like it has a full range of counts. This means that you may not be able to make a useful comparison between it and other frames. Using Normalize to Series preserves the relationship. Select Visual Type - For volumes, this allows you to select a view of the volume data. [various icons] [Rendering Modes] - This allows you to select a rendering mode from the pop-up list. See the section below on Rendering Modes. Threshold - For 3D volumes rendered using the ISO Surface, this adjusts the number of Hounsfield Units above which to create a rendered surface. Switch Image Orientation - This sets the orientation of the image. Select the appropriate orientation from the list. [various icons] [Volume type] - This allows you to select how much of a volume to display: a single slice; small, medium, or large slab; or the entire volume. Select Overlay - This toggles the active status of the top image (making it active for setting the colormap, for example). This is available only for fusion studies. Select Underlay - This toggles the active status of the bottom image (making it active for setting the colormap, for example). This is available only for fusion studies. [no icon] [no icon] [no icon] Sum - This displays a composite image in which each pixel is the sum of the corresponding pixel in all the images. Max - This displays a composite image in which each pixel is the max of the corresponding pixel in all the images. Isocontour - This displays a pop-up menu of all the isocontour ROIs in the image, and a slider to control the sensitivity of the edge detection of the object. The slider settings for the ROIs are saved as part of a Preference. Here are the options for the Scroll feature above: Scroll forward Scroll backward Click-drag down Click-drag up Philips Healthcare Right arrow key Page Down key Left arrow key Page Up key 22 General Information NM Application Suite Release 1.0

32 Using the Control Panel 2.4 Scroll forward Scroll mouse wheel backward End key (scrolls forward to last frame) Scroll backward Scroll mouse wheel forward Home key (scrolls back to first frame) Using the Movie Controls The Movie tool displays a panel at the bottom of the viewport with these features: 1. Play/Pause Figure 11 Movie controls 2. Change Frame Rate - This slider controls the speed at which the cine plays (1-100 frames per second) Using the Viewer Label Editor Clicking Viewer Label displays a panel at the bottom of the viewport with these features: Philips Healthcare Rev A Figure 12 Viewer label editor 1. Text - Type label text here. 2. Location - You can either enter coordinates in the X and Y boxes or drag the label directly in the viewer. When you drag, the location values update automatically. 3. Create - Click this to place the label in the viewer. 4. Update - Click this to apply any edited values to the selected label. 5. Label list - This is a list of all the labels in the active viewer. 6. Delete - This deletes the selected label. 7. All Frames - Checking this displays the label on all frames of a cine. An easy way to use the Viewer Label Editor is to just click Create, and then edit and move the new label directly on-screen by typing in the label box or dragging it. You can edit and move an existing label similarly. NM Application Suite Release 1.0 General Information 23

33 2.4 Using the Control Panel Labels are also available as Preferences. In the Preference Editor, you can click on the Labels tab and see labels for all images, sorted by workstep. You can edit the text and location within the editor. For more on creating and editing Preferences, see the section on Creating and Editing Preferences on page 43. Rendering Modes MaxIP - For each pixel in the output image a ray is cast through the volume. At equally-spaced discrete sampling positions, the value is (tri-linearly) interpolated from the nearest volume samples/voxels. The output image pixel is assigned the maximum value sampled by the ray. This creates a 2D image of the brightest voxels. MinIP - This is the same as MaxIP, except each output image pixel is assigned the minimum value sampled by the ray. Average - This is the same as MaxIP, except each output image pixel is assigned the average value sampled by the ray. Summed Intensity - For each pixel in the output image a ray is cast through the volume. At equally-spaced discrete sampling positions, the value is (tri-linearly) interpolated from the nearest volume samples/voxels. The output image pixel is assigned the sum of all values sampled by the ray. Shaded Opacity Blend - For each pixel in the output image a ray is cast through the volume. At equally-spaced discrete sampling positions, the value is (tri-linearly) interpolated from the nearest volume samples/voxels. Each value along the ray is assigned a (RGBA) color by applying a color and opacity map. A color map transforms a scalar value to an RGB color. An opacity map transforms a scalar value to an opacity value. Each of the color values for the sampling positions are then composited together. Phong shading is applied and the final color value assigned to the output image pixel. Opacity Blend - This is the same as Shaded Opacity Blend, but without the shading. ISO Surface - The surface representing constant value within the volume. Using the Application Palette The Application Palette allows you to select an application with which to view a study. You can switch to any application, however if the study is not compatible with the application you cannot load any of the data. Since this Manager is always available, you can switch applications at any time. Philips Healthcare 24 General Information NM Application Suite Release 1.0

34 Using the Control Panel 2.4 Figure 13 Application Palette Philips Healthcare Rev A Image and Curve Manipulation The Image and Curve Manipulation Tool (ICMT) allows you to perform processing operations on images and on generated curves. These operations can include (but are not limited to) image math, filtering, correction, and curve generation. You can perform multiple processing functions without having to store the processed image in the database and then recall the image to perform the next function. The ICMT can manipulate the following image types: Static Dynamic SPECT Gated SPECT Total Body Generated result datasets are DICOM 3.0 compliant, and do not overwrite existing data. NOTE: Whenever you create a new result to be used in another application, be sure to follow these guidelines: Always visually verify the dataset for artifacts, erroneous results, or any non-conformity. Always verify the dataset against the minimum dataset requirements for the application. IMPORTANT: Be sure to experiment with different values when performing operations on an image (math, filters, correction, etc.) to get the results you need. Do not count on any default to create an image that meets your specific needs. NM Application Suite Release 1.0 General Information 25

35 2.4 Using the Control Panel Using the ICMT The ICMT has 10 panels that allow you to perform different sorts of operations: Image Crop: This allows you to create an image that is a subset of another image. See the section on Image Crop on page 27 for details. Image Filter: This provides four common filters to use on images: Laplacian, Spatial 3x3, Spatial 5x5, and Spatial Temporal. See the section on Image Filter on page 27 for details. Image Math: This provides the four basic math operations (+,, x, /) for both image math and constant math. Nth Root and Clipping is also included. See the section on Performing Operations on Images on page 28 for details. Frame: This provides five functions that operate on frames: Append, Compress, Extract, Image Mask, and Matrix Convert. See the section on Manipulating Frames on page 30 for details. Image Orient: This provides tools for reorienting an image by rotating or flipping it, and by reversing the frame index. See the section on Image Orient on page 34 for details. Astonish: This provides functions to support Planar Astonish as a deconvolution image processing algorithm for optimized resolution recovery on non-gated planar static, planar dynamic, planar total body, and tomographic projection data sets. See the section on Astonish on page 35 for details. Attenuation Correction: This provides Chang s-based edge detection based on the isotope used. See the section on Attenuation Correction on page 36 for details. Motion Correction: This provides functions for performing manual motion correction. See the section on Motion Correction on page 37 for details. Create Curve: This provides tools for creating curves and calculating ratios based on ROIs. See the section on Create Curve on page 39 for details. Volume Knitting: This allows you to stitch reconstructed tomograms together to create a larger volume. See the section on Volume Knitting on page 42 for details. To use a panel, click on its arrow button: Philips Healthcare Figure 14 Button to open an ICMT panel 26 General Information NM Application Suite Release 1.0

36 Using the Control Panel 2.4 The Status Bar at the bottom of the viewing area displays information relating to the operation to be performed, for example, Select a region size, or Image type is not compatible with this operation. Check this box for quick instructions or related information. Figure 15 Status Bar (lower right) NOTE: Some ICMT panels use ROIs as inputs. You cannot use ROIs drawn in the Define Regions workstep as input. You must draw new ROIs using the Measurements feature in the Utilities Data Manager or in a viewport s context menu. In the ICMT, image names may not match what you see in a viewer or the EBW Patient Browser. This happens when there are multiple datasets with the same Object Name (for example LAT ). The Object Name is taken from the Image ID. The ICM Tool then uses the name as a base string and appends a number. So you could see names like LAT, LAT-1, LAT-2, etc. More technically, for multiple images below the same series, the series PIIM object is constructed once and is reused for all the subsequent images having the same Series Instance UID. Image Crop NOTE: This feature only works for whole body, static, dynamic, and tomo datasets. Philips Healthcare Rev A 1. Open the Image Crop panel using its arrow button. 2. Select a dataset from the Image pull-down menu. 3. Click Open Image ( ). A crop box with the dimensions of the Crop Size appears. 4. If necessary, change the crop size using the Crop Size pull-down menu. 5. Position the crop box by dragging it. Be sure to drag the lines that define the box, not the interior of the box. 6. Type a name for the result file in the Result text box. 7. Click Create ( ). Image Filter The result appears in a new tab in the current viewport. The filtering process is similar for static and dynamic images. Filtering is accomplished by assigning each pixel in an image, one at a time, a new count value. The new count value is determined by the specific filter NM Application Suite Release 1.0 General Information 27

37 2.4 Using the Control Panel weighting arrangement applied to the image. In other words, each pixel is multiplied by a number that is determined by the pixel s position relative to the center pixel. The products of these multiplications are added together, but only the pixel in the center is assigned the resulting weighted average count value. The filter is then moved and the process repeated, until every pixel in the image has been treated as the center and assigned a new, weighted average count value. For details on each filter s coefficients and weighting arrangement, see the section on Filters in the NM Application Suite Reference Manual. NOTE: Filtering an 8-bit deep image will convert it to a 16-bit deep image. 1. Open the Image Filter panel using its arrow button. 2. Select a dataset from the Image pull-down menu. 3. Select a Method. There are four filters available: Laplacian Spatial (3x3) Spatial (5x5) Spatial Temporal 4. Type a name for the result file in the Result text box field. 5. Click Create ( ). The result appears in a new tab in the current viewport. Performing Operations on Images Use the Image Math panel to perform image math. Arithmetic operations can be applied to these datasets: Two static Two dynamics One static and one dynamic When operating on two dynamic images, both images must have the same number of frames; the operation is applied between corresponding frames. In subtraction, for example, frame 1 of the second image set is subtracted from frame 1 of the first image set, frame 2 of the second image set is subtracted from frame 2 of the first image set, and so on, through all the frames of the two dynamic image sets. Philips Healthcare 28 General Information NM Application Suite Release 1.0

38 Using the Control Panel 2.4 Philips Healthcare Rev A If one image is dynamic and the other is static, then the operation is applied between each frame of the dynamic image and the static image. The resulting image is a dynamic image with the same number of frames as the dynamic image. You can operate on images with different matrix depths, but not on images with different matrix sizes. For example, an image with a matrix depth of 8 and an image with a matrix depth of 16 can be added together, however, the resulting image has a matrix depth of 16. On the other hand, an image with a matrix size of 64 x 64 cannot be added to an image with a matrix size of 128 x 128. The Matrix Convert function can be used to convert an image to a specified matrix size. Some image operations may produce results that at first may seem unexpected. For example, when multiplying an image, the max value for the image may change. This can happen if some pixels become saturated (reach a value of 32,767) as a result of the multiplication. In this case, each of the saturated pixels will have the maximum value. In Math operations, all header information from the source image is copied to the new result, so, for example, if you save the result image to the database you will not see the current date as the exam date. When performing image math, you must select an Operation and possibly the Type of information to operate on. Operations: Add: This adds the second operand to the image. Subtract: This subtracts the second operand from image. Multiply: This multiplies image by the second operand. Divide: This divides the image by the second operand Clip: This allows you to set a threshold or clip value by checking Clip Above or Clip Below. To threshold, type a value in the text box. To clip, type a value in the text box and check Set to 0. The difference is that Set to 0 sets values beyond the one in the text box to 0 instead of to the value in the text box. The value can be a number or a percentage, so you can type 1100 or 50%, for example. The percentage is calculated based on the image max. (The image minimum and maximum appear for reference below the text box.) Nth Root: This takes the Nth root of the image using a value for N that you set in the Nth Root box. Types of information: Image: The values of the pixels in a second image that you select (using the Image 2 list) are used for the second operand. Constant Value: The value you type in (using the Value box) is used for the second operand. The value can be a number or a percentage, so you can type 1100 or 50%, for example. The percentage is calcu- NM Application Suite Release 1.0 General Information 29

39 2.4 Using the Control Panel lated based on the Image max value. (The image minimum and maximum appear for reference above the Result box.) Image Statistics: The method you choose from the Method menu is used for the second operand: Minimum calculates the minimum pixel value (in all frames). Average calculates the mean pixel value (using the sum of all the frames means). Maximum calculates the maximum pixel value (in all frames). ROI Statistics: This allows you to select an ROI from the list in From ROI, and use its statistics as an operand. As in Image Statistics, you can select the Method (Minimum, Average, or Maximum). For multi-frame images, check Frame By Frame to apply the Method to each frame (rather than to all frames at once) to create the equivalent result frame. NOTE: You must use one or more ROIs created using the Measurement tools to use ROI Statistics, not ROIs drawn as part of the Define Regions workstep. Also, the list of available ROIs in the From ROI list consists of all ROIs for the image currently displayed in the Image menu, even if there are multiple instances of the image, in different containers. The basic procedure for using the Image Math panel is: 1. Open the Image Math panel using its arrow button. 2. Select a dataset from the Image pull-down menu. 3. Use the controls described above as appropriate. 4. Type a name for the result file in the Result text box field. 5. Click Create ( ). Manipulating Frames Image Frame allows you to manipulate frames by appending, compressing, extracting, masking, and converting the matrix. Frame Append Use Frame Append to add frames from additional acquired phases or statics to create a single continuous dataset. Images to be appended must have the same matrix size. NOTE: You can only append static or dynamic images. You can also append static images to dynamic datasets. In this case, the static images are appended to the last phase of the dynamic dataset. Also, for statics, the frame interval is the average of all the intervals. Philips Healthcare 30 General Information NM Application Suite Release 1.0

40 Using the Control Panel 2.4 You can append two or more static images to create a single-phase dynamic dataset, or two dynamic images to create a multi-phase dynamic dataset. Each phase maintains its original time per frame information. You can also append static images to the end of a dynamic dataset. In this case, the static images inherit the image rate of the dynamic dataset. NOTE: Images are appended using the time normalization method. For static images, each frame interval is the average of all frame intervals. The maximum pixel and total counts of each appended image are scaled up or down to match the image rate. NOTE: If you create a dynamic dataset from static images for use in the ICMT Motion Correction tool, you must save the data and exit the application before performing motion correction. 1. Open the Image Frame panel using its arrow button. 2. Select a dataset from the Image pull-down menu. 3. Select Frame Append from the Image Frame icon s pull-down menu. 4. Select a dataset to append from the Image 2 list. 5. Type a name for the result file in the Result text box. 6. Click Create ( ). Philips Healthcare Rev A Frame Compress Use Frame Compress to reduce the number of frames in a dynamic image set. This divides the original frames into small groups, each containing an equal number of frames, which you specify. The frames in each group are then added into one frame. If the number you specified does not divide evenly into the number of frames of the original dataset, the remaining frames are discarded. For example, if the number of frames of the original data is 16, and the number you specify is 2, then every two frames are added together to create an 8-frame dynamic image. But if the number of frames of the original data is 16, and you specified 3, then the result is a 5-frame dynamic image, and the sixteenth frame of the original dataset is discarded. For multiphase datasets, the number of frames available to compress is the number of frames in the smallest dataset. You cannot compress static or whole body datasets. NOTE: In some cases, when multiple frames of a 16-bit image are added together, a condition called rollover occurs. Rollover occurs when the pixel value exceeds , thus rolling over to a pixel value of 0. This causes the resulting image to have black holes where rollover occurred. If this is not the desired effect, scale down the pixel values by multiplying the image by a small value (such as 0.05). NM Application Suite Release 1.0 General Information 31

41 2.4 Using the Control Panel One situation where you might use Frame Compress is when you want to combine a fast phase dynamic with a slow phase dynamic. In a case like this, you could compress the first phase of 180 frames of 1 sec to 9 frames of 20 sec in order to then combine them with a second at 20 sec/frame. The resulting dataset is a single phase dynamic, since the frame rates of the 2 dynamics are now the same. For gated tomo or reconstructed gated tomo data, you cannot select the number of frames to compress: all bins are compressed, yielding just a tomo or reconstructed tomo (ungated) dataset. In this case the controls are replaced by the text Compress All Bins. 1. Open the Image Frame panel using its arrow button. 2. Select a dataset from the Image pull-down menu. 3. Select Frame Compress from the Image Frame icon s pull-down menu. 4. Set the Compress By Factor by typing a number or using the arrow buttons to increase or decrease the number. Frames are compressed in groups (see the illustration below). Setting the Compress By Factor to 5 compresses 5 frames at a time. By default, each frame is in exactly one group of compressed frames. However, you can create an overlap or skip frames using the Offset feature. This specifies the number of frames to skip, starting from the first frame in a group, before starting a new group. Figure 16 Using Compress By Factor and Offset For gated planar datasets, you can either compress all bins to one bin, or compress by a factor of 8. So if you have 24 bins, you can compress to 1 bin or to 3. NOTE: If you need to use an Offset that is different from the default (which is the same as the Compress By Factor value), use the arrow buttons to increase or decrease the existing number. 5. If necessary, adjust the Offset. Philips Healthcare 32 General Information NM Application Suite Release 1.0

42 Using the Control Panel If Compress By Factor is 5 and Offset is 5, compression is performed on frames 1-5, then frames 1-5 are skipped before starting a new group, 6-10; then the first 5 frames of the new group are skipped before starting another group, and so on. If Compress By Factor is 5 and Offset is 2, compression is performed on frames 1-5, then the first two frames are skipped and a new group of 5 starts on frame 3, then the first two frames of the new group are skipped and a new group of 5 starts on frame 5, and so on. Similarly, increasing the Offset to 7 would create a gap of 2 frames after every 5. (See the figure above.) 7. Type a name for the result file in the Result text box. 8. Click Create ( ). The result appears in a new tab in the current viewport. Extracting Frames Use Frame Extract to remove a frame or range of frames from a dynamic image. The new dataset is the extracted frame(s). NOTE: You can extract frames from tomo, gated, dynamic, or reconstructed tomo datasets. NOTE: To extract frames from a tomo dataset, it must be a 360 tomo with 64 or 128 projections. To create a 180 tomo you can just extract half the frames. If you extract frames in the middle of a projection, you cannot save them as a new tomo, but you could create a static from them. To extract frames from a gated tomo dataset you must compress all the bins first (using the Frame Compress feature). 1. Open the Image Frame panel using its arrow button. 2. Select Frame Extract from the Image Frame icon s pull-down menu. 3. Select a dataset from the Image pull-down menu. 4. For multiphase datasets, select a Phase to extract. NOTE: For multiphase datasets, you cannot extract across phases. Philips Healthcare Rev A 5. Type the number of the start frame in the Extract Frame From box or use the arrow buttons to increase or decrease the existing number. 6. Type the number of frames to extract in the Count box or use the arrow buttons to increase or decrease the existing number. 7. Type a name for the result file in the Result text box. 8. Click Create ( ). The result appears in a new tab in the current viewport. NM Application Suite Release 1.0 General Information 33

43 2.4 Using the Control Panel Matrix Conversion Use Matrix Convert to convert an image frame from one matrix size to another. This function is intended for use in situations where data has been acquired in a matrix size that is incompatible with a desired processing function. All images are automatically converted to a depth of 16; 16-bit deep images cannot be converted to 8-bit deep images. Total counts of the new image are within 5% of the original image. 1. Open the Frame panel using its arrow button. 2. Select a dataset from the Image pull-down menu. 3. Select Matrix Convert from the Image Frame icon drop-down menu. 4. Select a matrix size from the Convert to drop-down menu. 5. To avoid averaging counts when converting, click Preserve Total Counts. This preserves the counts in each pixel. 6. Type a name for the result file in the Result text box. 7. Click Create ( ). Image Mask The result appears in a new tab in the current viewport. Use the Mask feature to set pixels inside or outside an ROI to Open the Image Frame panel using its arrow button. 2. Select a dataset from the Image pull-down menu. 3. Draw an ROI to use as a mask (use one of the Measurement tools in the context menu available when you right-click on an image). 4. Click Mask In to set pixels inside the ROI to 0; click Mask Out to set pixels outside the ROI to Select the ROI to use as the mask from the From ROI list. 6. Type a name for the result file in the Result text box. 7. Click Create ( ). The result appears in a new tab in the current viewport. Image Orient For an image, you can rotate and flip to change the orientation; for a cine, you can reverse the order of the frames. This feature works on all datasets except for reconstructed gated tomo datasets. Philips Healthcare 34 General Information NM Application Suite Release 1.0

44 Using the Control Panel 2.4 NOTE: The Image Tools Data Manager contains tools to rotate and flip images. The difference between those tools and the ones in this panel is that the former only change the display; the ones in this panel create new image data, which you can then save. 1. Open the Image Orient panel using its arrow button. 2. Select a dataset from the Image pull-down menu. 3. Select an orientation from the Rotation menu. NOTE: The Reverse Frame Index menu item reverses the order of a set of frames. 4. Type a name for the result file in the Result text box. 5. Click Create ( ). The result appears in a new tab in the current viewport. Astonish Astonish is an image restoration process (a special type of filter) to enhance the image resolution for a planar image by using the Blind Deconvolution technique. Image restoration is a process to remove or minimize known degradations in an image. You can use any data except for volume and tomo. The default radioisotope is the one used to generate the image, but you can override it. The default isotopes are: Au-195m Co-57 Co-58 Co-60 Cu- 62 F-18 Fe-59 Ga-67 Ga-68 I-123 I-125 I-131 In-111 Kr-79 Kr-81m MO-99 Tc-99m Tl-201 Xe-127 Xe-133 Philips Healthcare Rev A Yb Open the Astonish panel using its arrow button. 2. Select a dataset from the Image pull-down menu. If the image has radioisotope, collimator, and pixel size information in the header, the Radioisotope, Collimator, and Pixel Size fields are filled in automatically (the Collimator and Pixel Size fields are for display only, not input). 3. If the Radioisotope field is not filled in, select a radioisotope from the pull-down list. NM Application Suite Release 1.0 General Information 35

45 2.4 Using the Control Panel 4. Set the number of Iterations. The higher the Iterations value is, the sharper the image becomes. However, at some point artifacts (noise) may start to appear. Be careful to strike a balance between sharpness and avoiding artifacts. 5. Type a name for the result file in the Result text box. 6. Click Create ( ). The result appears in a new tab in the current viewport. Attenuation Correction You can use Attenuation Correction on an image to compensate for the absorption and scattering that occurs in count information as it passes from inside the body to the detector. The attenuation correction method used is Chang's Attenuation Method. NOTE: Attenuation correction is only appropriate for transverse images, and operates only on reconstructed tomo datasets. The algorithm estimates the attenuation at any point by computing (e µd ), where µ is the amount of attenuation per cm, and d is the distance from any point to the edge of the body in cm. This is a post reconstruction correction algorithm. For transverse brain images, edge detection is automatic. For other transverse reconstructed images, for example cardiac, you can manually define ROIs and then apply the attenuation correction. Attenuation Correction requires the following data: Transverse images in 64 x 64 or 128 x 128 matrices An attenuation coefficient based on the radionuclide used (automatically extracted from the patient header) A pixel calibration factor (automatically extracted from the patient header) The attenuation coefficient is determined by the radioisotope used to generate the image, but you can override that value. The default attenuation coefficients are: Isotope Energy (kev) Coefficient Au-195m Co Ga-67 90,184, Philips Healthcare I I General Information NM Application Suite Release 1.0

46 Using the Control Panel 2.4 Isotope Energy (kev) Coefficient In , Tc-99m Tl , Xe Open the Attenuation Correction panel using its arrow button. 2. Select an image from the Image menu, then click Open ( ). 3. Click Identically in all Frames to draw an ROI the same size in all frames; click Individually Frame by Frame to draw an ROI in each frame based on edge detection within that frame. 4. Resize and readjust the ellipses, if necessary. 5. If the isotope is wrong, select the correct one from the Isotope pull-down menu. The default Coefficient is filled in according to the selected Isotope. If you need to redo the edge detection, click Reset. 6. Type a name for the result file in the Result text box. 7. Click Create ( ). The result appears in a new tab in the current viewport. Philips Healthcare Rev A References 1. Budinger TF, Gullberg RT, Nuesman RH: Emission computer tomography. Image Reconstruction from Projections: Implementation and Applications, ed G. T. Herman, Springer-Verlag. 2. Budinger TF, Gullberg RT: The use of filtering methods to compensate for constant attenuation single-photon emission computer tomography. IEEE, Transactions in Biomedical Engineering BME-28: Chang LT: A method for attenuation correction in radionuclide computer tomography. IEEE, Transactions in Nuclear Science NW-25: Motion Correction Use Motion Correction to adjust the positions of the loaded images to correct for patient motion. When you open a dataset for motion correction, you see a splash display so you can interactively browse the images in the study. You can only motion correct gated planar and dynamic datasets. You can have the system calculate corrected positions automatically, or you can manually correct images yourself. NM Application Suite Release 1.0 General Information 37

47 2.4 Using the Control Panel 1. Open the Motion Correction panel using its arrow button. 2. Select a dataset from the Image pull-down menu. 3. Click Open ( ). The dataset opens as a splash display in a new viewer. 4. Click a Correction Method. NOTE: Marker boxes are not used to locate significant image areas; they are just for reference. You could use the measurement tools to draw reference objects as well. 5. In the Markers section, click the button representing the number of markers to use. Figure 17 A marker box 6. Drag any of the marker boxes and center them on significant areas within the image. 7. Adjust their size to constrain the area in which the software looks for the image markers. 8. Scroll through all the images and make sure that the marker boxes enclose the significant area in each image. If they do not, resize the image. 9. If there are image markers you do not want to appear in the preview, check Blank Regions. 10. To avoid accidently moving or resizing markers once they are in place, click Lock Markers. 11. Examine the frames. If you selected Manual, you can adjust them by dragging or typing values in the X and Y fields in Frame Offset by Axis. 12. If you have moved any images and want to reset them to their original positions, click Reset. This resets all adjusted images simultaneously. You can only reset the markers and images on the original image, not on a Result image. 13. Type a name for the result file in the Result text box. 14. Click Create ( ). Depending on the placement of the marker boxes, the motion correction algorithm may produce artifacts. The only way to fix an artifact Philips Healthcare 38 General Information NM Application Suite Release 1.0

48 Using the Control Panel 2.4 problem is to try placing the marker boxes differently and redoing the correction. The result appears in a new tab in the current viewport. Create Curve You can use the Create Curve panel to create curves and ratios for ROIs, and to export an ROI to an Excel file. Curves appear in a new viewer. When a curve viewer is active, the ICMT Data Manager has panels that allow you to perform operations on curves instead of images. To switch from the curve functions to the image functions, click on the tab of an image viewer. Clicking on the tab of a curve viewer switches back. IMPORTANT: You cannot use ROIs created in the Define Regions workstep. You can only use ROIs you draw using the Measurements tool, available in the Utilities Data Manager or the right-click context menu. You can create ROIs at any time, not just when the Create Curve panel is active. NOTE: If you create a curve from a multiphase dataset, it displays frame data, not time data. This means that the x axis displays frame numbers, not time increments. Philips Healthcare Rev A ROIs you create appear in the ROI pull-down menu. 1. If you do not already have an ROI, create one using one of the Measurement tools available in the Utilities Data Manager or context menu. 2. Select an ROI using the ROI dropdown box, or click Select All to select all the ROIs. Select All selects all ROIs in all containers. However, it usually only makes sense to create curves for ROIs in one viewer. For this reason, be sure the ROIs you have are all confined to one viewer. 3. Select a method from the Method drop-down menu. The methods are: Min plots the minimum pixel value for each frame. Mean plots the mean pixel value for each frame. Max plots the maximum pixel value for each frame. Sum plots the total pixel values for each frame. 4. Click Export ROI if you would like to export the ROI information to a comma-separated (.csv) Excel file. Using the Save As dialog, this exports only the selected ROI. 5. Click Create ( ). NM Application Suite Release 1.0 General Information 39

49 2.4 Using the Control Panel 6. In the Curve field, type a name for the curve to be plotted. This is the label for the curve that appears in the plot. 7. In the Result field, type a new name for the plot or select an existing plot from the pull-down menu. This is the name that appears in the resulting viewer s tab. A new viewer appears with the curves in it, and the ICMT panel changes to display a new set of four panels. See the following sections for more on these. Curve Filter: This provides four ways to interpolate points on a curve: Smooth Filter (3- and 5-point), Linear Fit, and Exponential Fit. Curve Math: This provides the four basic math operations (+,, x, /) for both image math and constant math. You can also use this to display a selected curve s information from a subset of its frames. Curve Decay Correction: This allows you to perform decay correction based on the isotope used in the study. Curve Properties: This allows you to export curve information, change the curve line color, and delete curves. NOTE: If a dataset has time information, you can change the X axis of the plot to display minutes, seconds, milliseconds, or frames. Use the mouse scroll wheel to cycle through the choices. Interpolating Curve Points (Curve Filter) Use this panel to interpolate points on a curve. 1. With a curve viewer active, click the Curve Filter arrow button in the ICMT panel. 2. Select a curve using the Curve list. 3. Select a Method: Smooth (5 point) Filter Smooth (3 point) Filter Exponential fit Linear Fit 4. In the Result Curve field, type a name for the curve to be plotted or use the drop-down button. This curve label appears in the plot. 5. In the Plot field, type a new name for the plot or select an existing plot from the drop-down menu. This is the name that appears in the resulting viewer s tab. 6. Click the Create icon. Philips Healthcare 40 General Information NM Application Suite Release 1.0

50 Using the Control Panel 2.4 Philips Healthcare Rev A Performing Math Operations on Curves When performing curve math, select an operation and optionally a second operand. The operations are: Add to Curve: This adds the second operand to the curve. Subtract from Curve: This subtracts the second operand from the curve. Multiply Curve: This multiplies the curve by the second operand. Divide Curve: This divides the curve by the second operand. Extract from Curve: This displays a selected curve s information from a subset of its frames. Operand Types are: Second Operand is Curve: The values of the pixels in a second curve that you select are used for the second operand. Second Operand is Constant Value: A value you type in is used for the second operand. To preform curve math: 1. With a curve viewer active, open the Curve Math panel using its arrow button. 2. Select a curve from the Curve 1 drop-down menu. 3. Select an Operation and an Operand from the pull-down lists. 4. Select a second curve from the Curve 2 menu. 5. In the Curve field, type a name for the curve to be plotted or use the drop-down menu. This curve label appears in the plot. 6. In the Plot field, type a new name for the plot or select an existing plot from the drop-down menu. This is the name that appears in the resulting viewer s tab. 7. Click Create. The result appears in a new tab in the current viewport. Decay Correction on Curves Decay Correction allows you to perform decay correction based on the isotope (and its half-life) used in the study. The default radioisotope is the one used to generate the image, but you can override it. The default isotopes are: NM Application Suite Release 1.0 General Information 41

51 2.4 Using the Control Panel Au-195m Co-57 Co-58 Co-60 Cu- 62 F-18 Fe-59 Ga-67 Ga-68 I-123 I-125 I-131 In-111 Kr-79 Kr-81m MO-99 Tc-99m Tl-201 Xe-127 Xe-133 Yb With a curve viewer active, open the Decay Correction panel using its arrow button. 2. Select a curve from the Curve drop-down menu. 3. Select a radioisotope from the Radioisotope drop-down menu. 4. In the lower Curve field, type a name for the curve to be plotted or use the drop-down menu. This curve label appears in the plot. 5. In the Plot field, type a new name for the plot or select an existing plot from the drop-down menu. This is the name that appears in the resulting viewer s tab. 6. Click Create. The result appears in a new tab in the current viewport. Curve Properties Curve Properties allows you to export curve information, change a curve s line color, and delete curves. To display statistics, select a curve in the Curve list and click Statistics. A panel appears at the bottom of the image, and contains counts, times, slope, etc. To change the line color, select a curve in the Curve list and click Line Color. This brings up the Windows color picker. Select a color and click OK. To export curve values, select a curve in the Curve list and click Export Curve. This brings up the Windows Save dialog. Use the dialog to name and locate the saved file. NOTE: Saved curve files are in the.csv (comma-separated values) format. To remove a curve, select a curve in the Curve list and click Remove. Volume Knitting Philips Healthcare You can use the Volume Knitting panel to stitch together multiple reconstructed tomographic volumes into a single output tomographic volume. For example, you could use this if you acquired multiple single 42 General Information NM Application Suite Release 1.0

52 Using the Control Panel 2.4 field of view (FOV) tomograms, reconstructed them and then wanted to create a single reconstructed tomogram with all of the FOVs knitted together. NOTE: This panel is only available in AutoSPECT Pro. 1. Use the Segment menus to load the images to be knitted. You can view the image by clicking on its Open button ( ). 2. Specify SPECT or CT using the radio buttons. 3. Type a name for the result in the Result box (or leave the default). 4. Click Create ( ). Creating and Editing Preferences This section describes creating and editing preferences for all applications except AutoSPECT Pro and the Review application. For information on setting preferences for AutoSPECT Pro see the section on Setting Preferences for AutoSPECT Pro on page 101. The Review application does not have any Preferences. You can create a Preference from scratch or create one using an existing Preference as a starting point. You can also update an active Preference with a single click. In all cases, you begin by modifying layouts, ROIs, and any application-specific parameters. Then you can create, modify, or update the Preference. For information on editing layouts, see the section on Editing Layouts on page 69. Use these controls for Preferences: Figure 18 Preference function buttons Philips Healthcare Rev A 1. Create preference 2. Open Preference Editor 3. Update to Current and Save 4. Show Factory Use the Show Factory checkbox to toggle the display of the default factory Preferences. The rest of the buttons are discussed in the following sections. NM Application Suite Release 1.0 General Information 43

53 2.4 Using the Control Panel Figure 19 Preference Editor 1. User Preferences 2. Factory Preferences 3. Preference tabs For new Preferences, there is a Custom Display Method. Using this in a Preference allows you to add up to 25 images to be displayed in the Custom Display layout in the Review workstep. (And if you use AutoMatching, the buckets can be filled automatically.) If you create a new Preference with only the Custom Display Method, the workflow will jump from the Setup workstep directly to the Review workstep, with no region definition or results worksteps. This allows you to create a Preference that is set up to display the bucketed data in very specific ways. For example, you could set the number of viewers and their sizes, add viewer labels, and set the colormaps so that whatever you load is displayed consistently each time without having to process anything. Creating a Preference Based on an Existing One In most cases, you will probably only need to make variants of existing Preferences. 1. Apply the Preference to use as the starting point, or base Preference. 2. Make changes to layouts, ROIs, and application inputs as necessary. For information on changing layouts, see the section on Editing Layouts on page Select the Preferences Data Manager. Philips Healthcare 44 General Information NM Application Suite Release 1.0

54 Using the Control Panel Click Open Preference Editor. The Preference Editor window opens. 5. Make other changes in the Preference Editor as necessary. 6. Type a name for the preference in the Save As box. 7. Click Save. Creating a New Preference It is easier to create a Preference by starting with an existing one and editing it, but you can also create one from scratch. NOTE: When you create a new Preference, the layouts may not match those in the Factory equivalent of the Preference. Philips Healthcare Rev A 1. Select the Preferences Data Manager. 2. Click Create Preference. The Preferences Editor window opens (see Figure 19). 3. Check one or more methods in the list. 4. Click Proceed. 5. Make changes to the Organs tab as needed by selecting an ROI, changing its color, selection type, or drawing type. 6. Make changes to the Parameters tab as needed by editing the parameters. 7. Type a name for the Preference in the Save As box. 8. Click Save. 9. Click Close to close the window. 10. Apply the new Preference using the Apply layout button ( ). 11. Make additional changes to the workflow layouts as appropriate. For information on changing layouts, see the section on Editing Layouts on page In the Preferences Data Manager, click Update to Current and Save. Quick Update 1. Make the necessary changes to the workflow ROIs, layouts, and inputs. For information on changing layouts, see the section on Editing Layouts on page Select the Preferences Data Manager. 3. Click Update to Current and Save. NM Application Suite Release 1.0 General Information 45

55 2.4 Using the Control Panel Using the Global Image Tools There are two types of tools that are available in all applications. They organize activities in these categories: Image Tools: This provides basic image-related tools you can use in all applications. See the section on Using the Image Tools for details. Utilities: This contains miscellaneous tools, for example, ones that create tiles, control the display of data on images, and show hidden viewers. See the section on Using Utilities for details. Using the Image Tools The following tools are available in the Image Tools manager: Tool Description Image Control Bar (ICB)- This allows you to control dynamic range, colormaps, intensity, and pixel values. See the section below for details. Contrast Stretch - This resets the grayscale values of only the pixels in the viewport such that the lowest pixel value becomes 0 and the highest becomes 255. The rest of the pixels are scaled appropriately. This is most useful when you are zoomed into an image far enough to have some pixels beyond the edges of the viewport. Histogram Windowing - This resets the viewer contrast level to a reasonable value for data that does not have SUV values calculated. Use this tool to make a quick contrast adjustment when the values are far off normal. Invert Gray Level - This inverts the pixel values on the underlying image, so darker areas become lighter, and lighter areas become darker. Depending on the colormap applied to the image, you may see a different effect. This is not available in AutoSPECT Pro. NOTE: Using this reverses the effect of some of the colormap editor controls. See below for details. Pan - This moves the image within the viewer. The image follows the motion of the mouse. Zoom- This increases and decreases the apparent size of the image. Drag the mouse up to enlarge the image, down to shrink it. Flip - This flips the image about the horizontal axis. Mirror - This flips the image about the vertical axis. Philips Healthcare Rotate Clockwise - This rotates the image General Information NM Application Suite Release 1.0

56 Using the Control Panel 2.4 Tool Description Rotate Counter-Clockwise - This rotates the image -90. Select All Viewers - When this is Off, changes in the Image Control Bar (see below) apply to all image viewers (not curves, results, etc.). When this is On, changes apply only to viewers that are selected. By default, all image viewers are selected. Selected viewers have L-shaped borders in the corners. Ctrl-click a viewer to deselect or reselect it. Alpha Blending - For fusion studies (SPECT/CT, PET/CT, etc.), this controls the visibility of the two images. Dragging the mouse horizontally affects one image, and dragging it vertically affects the other. This is available only for fusion studies. Reset Image Viewing Settings - This undoes changes to all viewers made using the other Image Tools, resetting the viewers to the state they were in when the workstep began. This is true for all tools except Roll and Rotate, available in the right-click context menu. Save - This has two options: Save and Secondary Capture. See Saving Data below for details. Film Display - This captures the entire viewing area for film. Click on the Film button at the top of the window to use the film functions. See the chapter on FilmView in Volume 1 of the Extended Brilliance Workspace manual for details on this. CAUTION: Ensure that any hardcopy output that is to be used for interpretation is of diagnostic quality (printed on Codonics printers, for example). Non-diagnostic output may lead to misinterpretation. An example of non-diagnostic output is a JPEG file printed on a Windows printer. Report Display - This captures the entire viewing area for a report. Click on the Report button at the top of the window to use the reporting functions. See the chapter on Report in Volume 2 of the Extended Brilliance Workspace manual for details on this. CAUTION: Ensure that any hardcopy output that is to be used for interpretation is of diagnostic quality (printed on Codonics printers, for example). Non-diagnostic output may lead to misinterpretation. An example of non-diagnostic output is a JPEG file printed on a Windows printer. Philips Healthcare Rev A Using the Image Control Bar (ICB) The ICB allows you to control color maps, intensity, pixel values, and upper and lower levels for an image. For fusion studies, there are two ICBs. In this case, the top one is for the overlay image, and the bottom one for the underlay image (usually the SPECT image). Use the following features to control image colors and dynamic range: NM Application Suite Release 1.0 General Information 47

57 2.4 Using the Control Panel Figure 20 ICB controls 1. Sliders 2. Drop-down menu for Color Map, Intensity, and Pixel Values Sliders: These adjust the background and brightness (dynamic range) of all images that use the current color palette. All numbers are pixel values. The left (lower) slider controls the background, or lower threshold of the color map, below which pixel values are displayed as one color corresponding to the lowest value in the color table. Lowering this enhances the low-count image areas. The right (upper) slider controls the brightness, or upper threshold, above which pixel values are displayed as one color corresponding to the highest value in the color table. You can adjust the background and brightness together by dragging the section of the line between the numbers. Color Map: This is available in the drop-down menu at the right end of the color bar. Select a color map from the choices in the menu. CAUTION: If you create your own colormaps you should be familiar with digital image processing. Incorrect colormaps may lead to misdiagnosis. It is your responsibility to verify and validate the colormaps you create or edit. Philips is not responsible or liable for user-created colormaps. Philips Healthcare does not guarantee the accuracy of any edited or user-created colormap. Intensity: The Intensity cycle field changes the intensity of the displayed image. It is available in the drop-down menu at the right end of the color bar. The intensity and associated functions are listed in the table below. Philips Healthcare 48 General Information NM Application Suite Release 1.0

58 Using the Control Panel 2.4 Weight Function Log3 y = e 1/2.3 Log2 y = e 1/2.0 Log1 y = e 1/1.5 Linear y = x Exp1 y = e 1.5 Exp2 y = e 2.0 Use Log1, Log2, or Log3 when you want to enhance the lower count portions of an image. Use Exp1 or Exp2 when you want to enhance the higher count portions of an image. Figure 21 Intensity functions 1. Intensity Factor 2. Counts 3. Log 4. Linear 5. Exp Philips Healthcare Rev A Saving Data Pixel Values: This is available in the drop-down menu at the right end of the color bar. Select an upper bound for the slider range from the choices in the menu. The range is from 128 to The default is based on the pixel values in the current image. The Save utility has two options: Save and Secondary Capture. When you use Save, the Save Image dialog appears: NM Application Suite Release 1.0 General Information 49

59 2.4 Using the Control Panel Figure 22 Save Image dialog for 2D applications Figure 23 Save Image dialog for AutoSPECT Pro 1. To save an image, check its box and optionally edit its name by clicking on it and typing. 2. For AutoSPECT Pro, additionally select the tab for the settings used in the current Preference, the Thickness for the Transverse and Oblique data, and type a Technologist Name. NOTE: You cannot set Thickness for Sagittal or Coronal data. Philips Healthcare 3. Click Save to save the images to the EBW Patient Directory. (Be sure to refresh the browser to see the new images.) 50 General Information NM Application Suite Release 1.0

60 Using the Control Panel 2.4 When you use the Secondary Capture Utility, the Save Secondary Capture Dialog appears. 1. Use the Save As pull-down menu to select an image format. Single-Frame Secondary Capture JPEG Multi-Frame Secondary Capture AVI 2. Type a Description and choose the appropriate settings. 3. For multi-frame and AVI formats, you can check Frame and specify the Start and End frames, and the number of frames to Skip. 4. For gated formats, you can check Bin and specify the Start and End bins, and the number of bins to Skip. 5. Check RGB to save color images, or Grayscale for grayscale images. 6. Click Save to save the images to the EBW Patient Directory. (Be sure to refresh the browser to see the new images.) CAUTION: Do not use captured images or saved screens for diagnostic use. They may not include all the information necessary for a diagnosis. For this reason, they are only for reference, or inclusion in documents such as reports and presentations. Using Utilities You can use the following tools to control the layout of the viewers: Tool Description Philips Healthcare Rev A Text Box - This allows you to type text on the image. To use it, click in the viewer and type. You can then move it by dragging, and set its properties (see Applying Properties Changes to an Object on page 59 for details). You can append the text, and by deleting text from the end backwards, you can replace text. Right-clicking on it displays a menu of options you can use for the object. This text is not saved; to preserve it you must create a secondary capture. To create text you can save, you must create a Viewer Label, available in a viewer s context menu. You cannot use this on MIP images. CAUTION: When you add annotations, be careful not to accidentally cover some patient information. If you do, you could misidentify a patient, potentially causing misdiagnosis. Arrow + TextBox - This is like the Text Box above, but it creates an arrow with text attached to it. To use it, click in the viewer where the arrowhead should be, then click to set the other end of the arrow. Now you can type the text. You can move it by dragging, and set its properties. Right-clicking on it displays a menu of options you can use for the object. You cannot use this on MIP images. CAUTION: When you add annotations, be careful not to accidentally cover some patient information. If you do, you could misidentify a patient, potentially causing misdiagnosis. NM Application Suite Release 1.0 General Information 51

61 2.4 Using the Control Panel Tool Description Measurements - This is a list of measurement tools. See Making Measurements below for details. NOTE: You cannot use this on MIP images. Image Information - This toggles the display of various image overlays. See Information Display (Image Information) below for details. Annotation Editor - This displays a dialog box that allows you to create and edit text annotations on images. See the section below on Creating Annotation Templates for details. Create Tiles - This tiles the viewing area, allowing you to add space for more viewers, for example. The viewers always fill the space in the viewer area. This means that if you reduce the number of viewers, the remaining viewers will grow to fill the space. Show Hidden Viewers - If you have closed a viewer in the current layout, it will be in this list. To display it again, select it from the list. Deidentify - This toggles the display of patient information on all images in the viewing area. Making Measurements There are many tools you can use to take measurements or draw ROI objects in a dataset. All the objects created by the tools have some properties in common, and they are available in the menu that appears when you right-click on the object: Duplicate - This creates a copy of the selected object in a different color. Properties - This displays the Display Properties panel at the bottom of the viewer, which allows you to change properties of the selected object, such as color, size, transparency, etc. See Applying Properties Changes to an Object on page 59 for details. Copy - This copies the selected object. To paste, click Paste on the toolbar. If you paste in the same viewer, the object is pasted on top of the original, from where you can move it. If you paste it in a different viewer it is pasted in an equivalent position, taking into account anatomical features in the image. Delete - This deletes the object. Cut - This removes the object but keeps it in memory, available for pasting. Text Label - This allows you to add text to the shape. Click where you want the text to appear, and type the text. You can move a text label anywhere on the image. When you move the cursor over it, it is high- Philips Healthcare 52 General Information NM Application Suite Release 1.0

62 Using the Control Panel 2.4 Philips Healthcare Rev A lighted, and displays a line connecting it to the object it is associated with. Properties - This brings up the Properties dialog for the shape (see Applying Properties Changes to an Object for more). Duplicate - This creates a copy of the selected object on top of the object. This means that it looks like nothing has happened. However, you can then move the copy to separate the two. Additionally, some measurement tools allow you to create other objects related to the tool. For example, if you have two points selected, you can create a line that uses the points. These secondary objects have the same properties and menu choices as the tools of the same name: when you right-click on the line, the menu for a line appears. Depending on the object, you may also see these options: Area - This calculates the area of an enclosed shape. Perimeter - This calculates the length of the edge of an enclosed shape (not the area of the shape). Length - For lines, this displays the line length. Mean Pixel Value - This displays the mean pixel value of all the pixels on the object s line. Maximum Pixel Value - This displays the value of the brightest pixel on the line. Minimum Pixel Value - This displays the value of the dimmest pixel on the line. Total Value - This displays the value of the sum of all the pixels along a line. Pixel Value Standard Deviation - This displays the standard deviation of the values of the selected pixels. Histogram - This displays a histogram of the pixels in an ROI. Profile - For lines, this displays a graph that describes the pixel value at each pixel on the line. Mirror Polyline On Point - This mirrors the ROI about a point that you define by clicking. Mirror Polyline On Line - This mirrors the ROI about a line that you define by clicking twice to define the endpoints. If you select two ROIs that enclose an area (circle, box, etc.), right-clicking on one of them allows you to select ROI Ratio from the menu. This displays the mean, min, max, and sum values for the two areas. NOTE: You cannot draw on a cine in progress. You must stop the cine before drawing. NM Application Suite Release 1.0 General Information 53

63 2.4 Using the Control Panel The following table describes all the measurement tools. Tool Description Point - This creates a point. Click to create the point. If you drag the point, the values update continuously. Line - This draws a line between two points. Click on the starting point, then on the end point. Angle - This allows you to draw two joined lines, which can be placed along two image features to measure the angle between them. Click on the starting point for the first line. Drag the cursor to define the line, and click again to set the vertex of the angle. Move the cursor to the end point of the second line and click to complete the angle. Open Angle - This allows you to draw two separate lines, which can be placed along two image features to measure the angle between them. Click where you want to start drawing your first line. Click at the end point of the first line to end the line. Click where you want to start drawing your second line. Move the cursor to the end point of the second line and click to complete the angle. Polyline - Draws a line with corners at set points. Click at the starting point of your line. Drag the mouse to the point where you want to change the line s direction. Click to set a corner. Draw other line segments in the same way. Double click at the end point of the line. Smoothed Polyline - Draws a line with curves at set points. Click at the starting point of your line. Drag the mouse to the point where you want to change the line s direction. Click to set a curve. Draw other line segments in the same way. Double click at the end point of the line. Circle - This allows you to draw a circle. To use, click to set a starting point for the circle. Drag the mouse until the circle is the correct size. Click again to complete the circle. Ellipse - This allows you to draw an ellipse. Click-drag to create the major axis of the ellipse. When the axis line reaches the right length, release the mouse. Move the cursor perpendicular to the first line until the ellipse is the desired size, and click to set the ellipse. Box - This allows you to draw a box. Click to set a starting point for the box. Drag the mouse until the box is the correct size and shape. Click again to complete the box. Rectangle - This allows you to draw a rectangle. Click-drag to draw one side of the rectangle. Move the cursor perpendicular to the line to expand the rectangle. Click to complete the rectangle. You can rotate or resize the rectangle by selecting it, clicking on a handle on the rectangle s corner, and dragging that corner to a new location. The entire rectangle will rotate or change size. Polygon - This allows you to draw a multi-point polygon. Click to set a starting point. Drag the cursor to create a side, and click to set a corner. Double-click to set the final corner and complete the polygon. Philips Healthcare Smooth Polygon - This allows you to draw a multi-point polygon with curved corners. Click to set a starting point. Drag the cursor to create a segment, and click to set a corner. Double-click to set the final corner and complete the polygon. 54 General Information NM Application Suite Release 1.0

64 Using the Control Panel 2.4 Tool Description Closed Contour - This allows you to draw a curved contour. Click to set a starting point for the curve. As you drag the cursor, a line appears along the its path. Click again to complete the shape. Closed contours can be edited by selecting the contour and drawing a path that starts and ends on the contour. The new path replaces the contour segment between those points. NOTE: If you draw a contour that loops over itself, the area inside any loop is ignored in reporting or calculating pixel values in the contour. Information Display (Image Information) This provides menus that control various kinds of annotations on the image. There are annotation templates that control which DICOM fields appear on images. These are set for Minimal, Normal, and Extended. You can edit the templates using the Annotation Template Editor. For more on this, see the section below on Creating Annotation Templates. Tool Description Normal - This displays a medium amount of patient information. Minimal - This displays a small amount of patient information. Extended - This displays all available DICOM annotation information. None - This removes all DICOM annotation information. Rulers - This toggles the display of rulers. Philips Healthcare Rev A Grid - This toggles the display of a grid centered on the image. Color Scale - This toggles the display of a color scale on the right side of the image. Creating Annotation Templates You can customize the text annotation on images using the Annotation Template Editor, changing such features as the text s color, font, position, and so on. To open the Annotation Template Editor, click Annotation Editor. NM Application Suite Release 1.0 General Information 55

65 2.4 Using the Control Panel CAUTION: When adjusting or creating annotations, be sure not to obscure key patient information. If patient information is obscured, misdiagnosis may result. Figure 24 Annotation Template Editor When a viewer displays an image, the annotation on the image is determined by six criteria: Whether the image is original data or derived (which usually means a processed image) Whether the image is a primary image or a secondary capture The image type (gated, tomo, etc.) Whether the image is an emission or transmission image The modality of the image (PET, SPECT, etc.) The amount of annotation set by the Image Information control in the Utilities Data Manager (this is not a property of the image, but something you set in the application) You can set values for all of these criteria, in addition to adding or deleting text and changing its properties. To change the annotation used for a particular type of image, you must first know what kind it is. To find out: 1. Select the image in the lower list in the EBW patient directory. 2. Right-click and select View Dicom Info from the menu. This displays the DICOM Information window. Philips Healthcare 56 General Information NM Application Suite Release 1.0

66 Using the Control Panel Find the Image Type field (0008:0008). This field has information on the first four of the bullets above. 4. Find the Modality field (0008:0060). This is the fifth bullet above. Now you can set the correct values in the Annotation Template Editor: 1. Set the first five values: the DICOM Modality field corresponds to the Modality pull-down menu (see the figure below); the DICOM Image Type fields correspond to Value 1, Value 2, Value 3, and Value 4. Figure 25 DICOM field values 1. Modality: 0008: Value 1: 0008:0008 (first value) 3. Value 2: 0008:0008 (second value) 4. Value 3: 0008:0008 (third value) 5. Value 4: 0008:0008 (fourth value) NOTE: The Modality determines the attributes that appear in the list on the left. For combined modalities, attributes that are unique to one modality are color coded. Use the legend at the bottom to identify them. Use the Reset button in the upper right at any time to undo changes. Philips Healthcare Rev A 2. Select the annotation level using the Annotation pull-down menu at the top of the window; check Set Level As Default to have images use the Annotation level as the default. 3. Set the Render Mode if the annotation should apply to specific display options. 4. Select a color for annotations, as necessary, by clicking the Color For All Annotations checkbox and selecting a color from the drop-down menu. 5. To add an annotation, drag it from the list on the left to the image area; to remove an annotation, drag it from the image area to the list. 6. Position the annotation in the image. The image area is divided into a 3x3 grid (you can see this in Figure 24). To move an annotation into another area, drag and drop it into the area. NM Application Suite Release 1.0 General Information 57

67 2.4 Using the Control Panel 7. To edit an annotation s text properties, double-click on the annotation in the image area. This displays some controls on the right side of the window: Figure 26 Annotation text properties 1. Display Name 2. Delimiter 3. Justification 4. Font 5. Color 6. Viewer Cell 7. Both 8. Underlay 9. Overlay 8. Type a new Display Name, as necessary. 9. For annotations that use two attributes on the same row (the Matrix annotation, for example), type a Delimiter character or string to use to separate the strings. 10. Change the Justification, Font, and Color as necessary. The color overrides any color set by the Color For All Annotations control. 11. For dynamic images, check Viewer Cell to have the annotations appear only on the first frame instead of on every frame. 12. For a fusion image, select whether annotations appear on the Overlay, Underlay, or Both by using the For Fusion, Show controls. 13. Click OK to confirm the changes, or Cancel to ignore them. To load the factory default for the current settings, click the Load button in the upper right. To remove changes you have made to a template, click Undo Changes. Click OK to cancel. Click Save to save the annotation. Philips Healthcare 58 General Information NM Application Suite Release 1.0

68 Drawing ROIs 2.5 Click Delete Template to delete the template. Applying Properties Changes to an Object The Properties panel allows you to control the appearance of measurement ROIs and text. 1. Right-click on the ROI and select Properties from the menu. If the viewer is too narrow to show the full Properties panel, you may be able to resize the viewer until it fits. 2. Use the Text properties and Graphic properties tabs to make changes to the properties as needed. NOTE: The text properties apply only to the ROI name, not any statistics that may be showing. 3. To use the settings on all objects of the type selected, check Set as Defaults. To include the current selection, check Apply for selection as well. 4. Click Apply. Not all properties apply to every object. 5. To close the Properties panel, click the close button in the upper right corner of the panel. 2.5 Drawing ROIs Philips Healthcare Rev A Ordinarily, you draw ROIs in the Define Regions workstep. But you can also draw them on any image using the Measurements tool available in the right-click context menu or in the Utilities Data Manager. For more on this tool, see the section on Making Measurements on page 52. In the Define Regions workstep, ROIs are drawn in three ways: Some ROIs may be drawn automatically as the workstep initiates. You can manually draw ROIs, and manually redraw ROIs that were created automatically. Some ROIs may allow you to use an automatic Detect Region feature to create the ROI without having to draw it manually. The most common procedure for drawing ROIs is to find the image on which to create the ROI (the one with an instruction on it), choose a drawing tool to use, draw the ROI, and proceed to draw all the required ROIs similarly. The drawing tools work intuitively, though the Closed Contour tool bears some explanation. To use this tool, click and drag in the window. When you release the mouse button, a line segment automatically connects the NM Application Suite Release 1.0 General Information 59

69 2.5 Drawing ROIs point where you started to the point where the cursor is when you release the button. This means that the cursor does not necessarily need to describe a complete shape, since you can complete a shape by releasing the mouse button. The software helps you know what ROI to draw in three ways: The cursor may become a pencil in the window in which to do the drawing. An instruction appears in the window (for example, Draw Gall Bladder ). A pencil may appear in the ROI control of the ROI to be drawn. To redraw an ROI, click on the Draw Region eraser icon of the ROI control you want to redraw (#3 in Figure 28). This removes the ROI. Then redraw it. To delete all manually drawn ROIs, click Redraw Regions (#3 in Figure 27). This also toggles the display of manually drawn regions. Figure 27 Global ROI controls 1. Add User Defined Region 2. Load Templates 3. Redraw Regions 4. Detect All Regions 5. Set ROI Type for all regions Philips Healthcare 60 General Information NM Application Suite Release 1.0

70 Drawing ROIs 2.5 Figure 28 ROI drawing controls 1. Select ROI Color 2. Detection Type 3. Draw Region 4. Detect Region 5. Set ROI Type for this region Editing ROIs To edit an ROI, first select it. If it is a geometrical shape, use the anchor points to resize it; if it is a Closed Contour, start drawing on the line where you need to edit. If you need to edit an ROI that is inside another ROI, select the outside ROI and move it to a different location on the image where it does not overlap the ROI you wish to edit. Edit the ROI and then return the other ROI to its original position Reusing ROIs Philips Healthcare Rev A Changing the Edge Detection Type You can create ROIs that you can reuse whenever you run the application. You might use this to create an ROI for a study that will be performed again later on the same patient, for example. You can create templates from any ROIs. To create a template, right-click on an ROI and select Save as template. To use saved templates, click Load Templates (#2 in Figure 27). This draws all the templates you have created. There are four edge detection types available in the Detection Type pull-down menu (#2 in Figure 28): Manual: This uses the currently selected tool (#5 in the figures above). NM Application Suite Release 1.0 General Information 61

71 2.5 Drawing ROIs Isocontour: This draws an ellipse, which you can move and resize to enclose an ROI. For details see Using Isocontours below. This is not available for all regions. Automatic: This creates an isocontour with no hints from a preliminary ellipse, as the Isocontour type has. As soon as you select this, edge detection is performed, as long as all required ROIs are present. If the required ROIs are not present, nothing happens until they are. This is generally for creating background ROIs. This is not available for all regions. Template: If there is a template set for a region, it loads automatically if this is selected Using Isocontours The Isocontour tool draws an ellipse, which you can move and resize to enclose an ROI. To move the ellipse, drag it; to resize it, use the handles to change the shape. To create an isocontour within the ellipse, click the Detect Region icon (#4 in Figure 28). To draw all isocontours at once, click Detect All Regions (#4 in Figure 27). This also toggles the display of the isocontours. Clicking it alternately draws them and hides them Drawing Other ROI Shapes You can use any of the ROI tools to draw ROIs. Select a shape from the ROI type pull-down list (#5 in the figures above): Box Rectangle Polygon Closed Contour Circle Ellipse Smoothed Polygon Other ROI-related Tasks Other ROI-related tasks you can perform are: Move an ROI: click and drag the ROI s outline. Philips Healthcare 62 General Information NM Application Suite Release 1.0

72 Using Viewers 2.6 NOTE: If the ROI is small and has many control points, it may be difficult to grab a portion of the outline with no control points. If this is the case, zoom in. This will make everything larger, including the space between control points. Change the color of an ROI: click the arrow on the ROI s color swatch (#1 in Figure 28) and pick a new color from the pulldown menu. Create a custom ROI (#1 in Figure 27); see Creating a Custom ROI on page 63 below for details Creating a Custom ROI You can use custom ROIs to create regions to analyze that are not provided by default. 1. Click Add User Defined Region (#1 in Figure 27). A new ROI control is added at the bottom of the ROI list. 2. Type a name for the ROI in the text field. 3. Set the color using the color swatch pulldown menu. 4. Set the visibility of the ROI: a check mark in the checkbox shows the ROI; unchecking the box hides the ROI. 2.6 Using Viewers Philips Healthcare Rev A The NM Application Suite viewers allow you to view medical image data using optimized default protocols and frequently used viewing protocols, based on image modality and image types. A viewer is a tabbed window that displays a single dataset. There are different types of viewers to display different kinds of datasets (static, gated, orthogonal, etc.). When data is loaded, the software uses certain criteria to decide which viewer to use to display the data. Certain tools are available in a context menu available with a right-click for each type of viewer. Viewers are combined into layouts, which can contain multiple viewers and types. NM Application Suite Release 1.0 General Information 63

73 2.6 Using Viewers Viewer Components Figure 29 Viewer components A viewer has five main components: 1. Tab(s) 2. Title Bar 3. Hide button 4. Minimize/Maximize button 5. Image area 1. Tab(s) The viewer tab displays some information that is relevant to the contents of the viewer. This could be the layout name, the bucket name that the dataset was loaded in, the dataset name, etc., or a combination of these. A tool tip for the tab contains additional information. You can use the tab to reposition the viewer or to switch between images if they are stacked. See the section below on Moving and Resizing Viewers for more on this. When a viewer has multiple tabs, the tab for the active image display is highlighted. 2. Title bar You can maximize and restore the viewer by double clicking on the title bar. 3. Hide button Use the Hide button to remove a viewer from the layout. To show a viewer again, use the Show Hidden Viewers button in the Utilities Global Tool. Philips Healthcare 64 General Information NM Application Suite Release 1.0

74 Using Viewers Minimize/Maximize button Click Minimize/Maximize to enlarge a viewer or to return the image to its original size. NOTE: You can also double-click on the viewer s title bar to resize the viewer. 5. Image area The image area of the viewer contains the image. Also, if you right-click in the image area, a menu appears with a set of tools that depends on the type of viewer. See the section on Context Image Tools on page 19 for details. The image area contains any annotations present, such as patient name, date of birth, etc., and in the lower left, values for brightness and background, abbreviated LL (Lower Level) and UL (Upper Level). These are the values under the Image Control Bar in the Image Tools Global Tool Linking Viewers You can select multiple viewers so they respond together to tools that affect a viewer. These tools are the ones in the context menu available when you right-click in the viewer and their duplicates in the Data Managers. To select a viewer, Ctrl-click in its image area; add or subtract viewers by Ctrl-clicking in them. One viewer is the controlling viewer. This is indicated by blue brackets in the corners. Any actions performed in the controlling viewer are duplicated in the linked viewers. Linked viewers have white brackets. Philips Healthcare Rev A Figure Moving and Resizing Viewers Viewer brackets, controlling viewer on the left You can move a viewer by clicking on its tab and dragging. You can move a viewer to two places: an empty area of the layout (an area with no viewers), or to the title bar of another viewer, where it becomes another tab. NM Application Suite Release 1.0 General Information 65

75 2.7 Using Layouts When you drag a viewer, the cursor includes a small square. The square contains a black + when the cursor is in an area where you can drop the viewer, and a red x when it is not. If you drop the viewer in an empty area, the viewer is resized to fit in the area. If you drop it on the toolbar of another viewer, the viewers stack. When viewers are stacked, clicking on a tab brings that viewer to the front. Viewers are laid out on a grid. You can resize a viewer, but only by resizing the section of grid that it occupies. This means that other viewers may change shape as well. To resize a viewer, drag one of its edges. For more on changing a layout to adjust viewers, see the section on Editing Layouts on page Using a Curve Viewer In the Review Results workstep, curve viewers have unique functions that are not available in other viewers. These are available when you right-click in a curve viewer. Y Axis To Domain Range - This redraws the graph so the Y axis minimum value is 0. AutoScale Y - This redraws the graph so the Y axis minimum is the min value instead of 0. Time Axis To Domain Range - This redraws the graph so the X axis minimum value is 0. AutoScale X - This redraws the graph so the X axis minimum is the min value instead of 0. Export To Spreadsheet - Selecting this brings up a save dialog allowing you to save the values in a comma-separated list (.csv file). Axis In Sec - Selecting this displays the time axis in seconds. Axis In Min - Selecting this displays the time axis in minutes. Show Grids - When this is checked, a grid is drawn under the curves. 2.7 Using Layouts A layout consists of one or more viewers, of any type. When a study loads in the Review application (the default for many studies), all the images appear in a default layout appropriate to the data that is loaded. You can select a different layout after the images load. The layouts are designed to be clinically appropriate to facilitate review. However, you can edit a layout so it addresses specific needs, and save it as part of a Preference. For more on this, see the section on Creating and Editing Preferences on page 43. Philips Healthcare 66 General Information NM Application Suite Release 1.0

76 Using Layouts 2.7 NOTE: A layout s viewports try to maximize the size of images during loading. However, the success of this attempt depends on a number of factors, such as the image size, aspect ratio, matrix size, brightness, and viewport size, and so on. For this reason, image sizes may vary. If you have loaded a study in the viewer, you can select a layout that provides a set of viewers that displays the data to its best advantage. See the list of layouts below to decide which one suits your purposes. The available layouts depend on the type of study that is loaded. NOTE: When you load a multiphase dynamic dataset, the image contrast and window are assigned based on the image that contains the maximum pixel value. A dynamic phase viewport can appear empty. Use Contrast Stretch and Histogram Windowing to make more of the pixels visible. To switch to a different layout, click on one of the choices under the workflow controls. Layouts you have viewed have a check mark on the left. Layouts that are available for the current data have a light background; unavailable layouts have a dark background. The layouts that are available depend on the data that is present. Layouts called SC images display secondary capture images. Figure 31 Examples of available and unavailable layouts Philips Healthcare Rev A 1. Visited layout 2. Current layout 3. Unavailable layouts The Custom Display layout is available at the bottom of the list of factory layouts in the Review workstep for all applications except for AutoSPECT Pro and the General Review application. It displays up to 25 datasets in a study (depending on what is loaded in the Custom Display buckets), not just the data needed to run the application. NOTE: When loading whole body data pairs, you may need to pay attention to the DICOM Instance Number. For a single pair, the application displays them correctly no matter how you load the datasets. The following layouts are available in the Review application. See the sections on individual applications for other application-specific layouts. NM Exam Overview Planar Viewing Protocols Whole Body Display NM Application Suite Release 1.0 General Information 67

77 2.7 Using Layouts Whole Body Horizontal Display Whole Body with Spots Display Spots Display Whole Body Intensity Compare All Dynamic Display All Dynamic with Spots NM Planar comparison side by side NM Planar comparison - fusion Gated Planar Display Secondary Captures Overview SPECT and SPECT/CT Viewing Protocols One View One View Comparison (2 volumes) SPECT 3 View SPECT 3 View Composite (projection) SPECT 3 View Raw Composite SPECT 3 View Composite (3D MIP) SPECT 3 View Comparison SPECT 3 View Comparison Composite (projection) Orthogonal2D View CT Series Display DX Series Display Xray Series Display PET Series Display MR Series Display US Series Display MPR 2D Fusion Display (side-by-side) MPR 2D Fusion Display NM SPECT comparison side by side Multi-Segment Fusion display Whole body Fusion display Philips Healthcare 68 General Information NM Application Suite Release 1.0

78 Using Layouts Editing Layouts A layout consists of viewers in a grid of tiles. If you change the size or shape of a tile, since the total grid area is fixed, usually it will necessarily affect the size or shape of at least one other tile. However, you can change the number of tiles in the grid, which creates larger or smaller tiles for the viewers to fit into. What this means is that editing a layout is often a tradeoff between the sizes and shapes of the tiles (viewers) in the layout. NOTE: The tiles in a grid are not independent rectangles. They are merely the space between the grid lines. So when you adjust a tile, you adjust grid lines, which affects adjacent tiles. To change the number of tiles in a layout, click Create tiles in the Utilities Data Manager and select a tile layout. If a layout does not display the data optimally for your needs, you have two choices. If you can change a default layout so it is more useful, you can just edit that and save the changes as part of the Preference or as a new Preference. But if you need multiple new layouts, you can create new ones using the Save Layout As button. You can start from an existing layout or from the Custom Display layout, which is available at the bottom of the list of layouts in the Review workstep for all applications except for AutoSPECT Pro and the General Review application. Philips Healthcare Rev A Figure 32 Custom Display layout and Save Layout As button The Custom Display layout contains up to 25 datasets in a study, not just the data needed to run the application. This means you could create a comprehensive layout or set of layouts, or remove some viewers so you see only the data you want. To create a new layout: 1. Select a layout to edit (factory layout, user layout, or Custom Display layout). 2. Make changes to the layout by removing and resizing viewers, changing the number of tiles, changing colormaps, etc. 3. Click Save Layout As and type a name for the layout. The new layout is added at the bottom of the list. NM Application Suite Release 1.0 General Information 69

79 2.7 Using Layouts 4. If you made changes to the way buckets are automatched, test the layout by exiting the application and running it again. To edit a saved layout, make the changes and then use Save Layout As to resave the layout using the same name. To delete a saved custom layout, right click on the list entry and click the Delete menu item. NOTE: You cannot delete the layout that is currently active: you must first select another layout. In making changes to a layout you can: Rearrange viewers in the tiles Resize the tiles Additionally, in all applications except for AutoSPECT Pro, you can: Hide viewers (leaving empty tiles) Change the number of tiles in the grid NOTE: A viewer always completely fills a tile. So when you change the size or shape of a tile, you change the size or shape of the tile s viewer accordingly. To rearrange viewers, drag a viewer to an open tile. You may have to create an open tile by temporarily stacking two viewers. To stack viewers, drag the tab of one onto the title bar of another. To resize a tile, drag one of its edges. Dragging a side edge changes the width of the tile, as well as that of the adjacent tile; dragging a top or bottom edge changes the size of the whole row of tiles, as well as the adjacent row. NOTE: You can resize tiles while they contain viewers. To hide a viewer, click the viewer s Hide icon in the title bar. To show a hidden viewer, use the Show Hidden Viewers tool in the Utilities tab. Figure 33 Title bar Hide control To change the number of tiles in the grid, use the Create Tiles tool in the Utilities tab. If you need fewer tiles in a single row (to allow for wider viewers), you can resize a tile on the end so both its edges are at the edge of the viewing area, making it a zero-width tile. However, if you do this, be aware that the now-invisible end tile may get populated with a viewer, which will be a zero-width viewer. If this happens, resize the tile, move the viewer, and size the tile back again. Philips Healthcare 70 General Information NM Application Suite Release 1.0

80 Using Layouts Editing Layouts in AutoSPECT Pro AutoSPECT Pro handles layouts slightly differently from the other applications. First, you can only edit layouts in the Review workstep. After editing a layout, click Save under the list of layouts and type a name for the new layout. To delete a layout, launch the Preference Editor and check Delete Protocol. When you save the Preference, the layout is deleted. Philips Healthcare Rev A NM Application Suite Release 1.0 General Information 71

81 2.7 Using Layouts Philips Healthcare 72 General Information NM Application Suite Release 1.0

82 3 AutoSPECT Pro AutoSPECT Pro is a fast and accurate application that allows you to automatically or manually reconstruct one or more SPECT, gated SPECT, Total Body SPECT, Vantage SPECT, or SPECT-CT projection datasets. SPECT datasets that you can manually reconstruct include cardiac, bone, brain, liver, and other SPECT datasets. When reconstructing data, you can view three instances of the data and change filter settings on each one for comparison. NOTE: Vantage is the trade name for the Philips nonuniform attenuation correction option using a gadolinium transmission line source. NOTE: In this manual, CT-AC indicates that CT is used for attenuation correction. AC indicates generic attenuation correction. Philips Healthcare Rev A For Brightview XCT data, CT images are paired with projection data. The result is that when SPECT data is loaded, the CT data is automatically placed in the AC map bucket. If you have edited the DICOM information for the CT images (for example, the Series Description), the pairing is broken. When the data is no longer paired, it is handled the same way as data from machines that do not support pairing, which is often the case: automatching is attempted. If automatching fails, you can bucket data manually. The AutoSPECT Pro workflow differs from that of the other applications. It has the standard Setup workstep to start the process, and the Review workstep at the end. However, instead of the Define Regions and Results worksteps, AutoSPECT Pro has the following: AC Map Generation: This workstep is available when you load the appropriate data. It allows you to create or inspect an AC map for processing. Reconstruction: This allows you to reconstruct unprocessed datasets, including SPECT, gated SPECT, and Total Body SPECT datasets. Reorientation: This allows you to reorient transverse datasets. For cardiac studies, short axis, horizontal long axis, and vertical long axis datasets are automatically generated from the reoriented transverse datasets, and displayed. For all studies, you can use this workstep to manually reorient the reconstructed dataset. When a dataset loads in the Setup workstep, its energy window is checked against a default for the radioisotope used. If the window is absent or differs from the default, a dialog appears allowing you to select an isotope, use the default, or use the settings from the source image. NOTE: All worksteps appear in the workstep list, but whether you can use one may depend on the data and preference selected. NM Application Suite Release 1.0 AutoSPECT Pro 73

83 3.1 The AC Map Workstep NOTE: Images that are 256x256 pose special issues. First, this size is not supported for gated datasets. For other datasets, processing can take a long time due to memory demands. Additionally, for 256x256 Total Body SPECT data, you should process segments separately and knit them afterward, rather than using the AutoSPECT Pro AutoKnitting feature. One possible SPECT/CT workflow might include processing in AutoSPECT Pro, viewing fusion displays, and loading the results to AutoQuant. To ensure that this sort of workflow is possible, always start with the original CT data and not a preexisting ACMap (for example, one created in JETStream Workspace). If the CT data is not available, the workflow may not complete correctly. However, you could still use a preexisting ACMap to perform attenuation correction of the SPECT data. 3.1 The AC Map Workstep This workstep is available for some Preferences if the right data is loaded. It has four layouts, determined by the input data and Preference selected: Reviewing an AC Map A QC layout for reviewing an existing AC map using a CT-AC or Vantage preference CT-AC layout: This is available if you have loaded a CT image using one of the CT-AC preferences. Vantage AC layout: This is available for Vantage data if you have also loaded transmission projection data using one of the Vantage preferences. Chang s AC layout: This is available for any SPECT image using a Chang s AC preference. NOTE: For SPECT cardiac studies acquired from BrightView XCT systems using a 1-segment protocol, the CT FOV is 14.4 cm in the axial direction. For patients with large hearts, to ensure the robustness of attenuation correction, 3D scatter correction, and Astonish resolution recovery, the attenuation maps are automatically extended by ~20 mm at both ends in the axial direction by duplicating the original end slices of the attenuation map. Hence, the effective axial map FOV is ~18.4 cm. If you provide an AC map as input to AutoSPECT Pro, the AC Map workstep displays the map in a splash format so you can review it. Philips Healthcare 74 AutoSPECT Pro NM Application Suite Release 1.0

84 The AC Map Workstep Using CT-AC Data If you are processing a CT-AC dataset, a new data manager for performing registration tasks becomes available. Use the controls below in the Registration data manager to register the SPECT image relative to the CT image. Figure 34 CT-AC layout controls 1. Translation Offset 2. Rotation Angle 3. Translation Increment 4. Rotation Increment Tool Description Translate: This allows you to manually translate the SPECT image in any of the three orthogonal views by dragging the image. You can also use the Translation Offset parameters to specify offset values in X, Y, and Z. Philips Healthcare Rev A Incremental Translate: This allows you to translate the SPECT image in increments specified by the Translation Increment control (and set as defaults: see Preferences for the ACMap Workstep on page 104). To translate, click on the image. The image is divided into triangular quadrants defined by an imaginary X drawn from corner to corner (see Figure 35). Clicking in a quadrant moves the image in the direction of that quadrant (for example, clicking in the left quadrant moves the image left). You can also use the Translation Offset parameters to specify offset values in X, Y, and Z. Rotate: This allows you to manually rotate the SPECT image by dragging it in any of the three orthogonal views. You can also use the Rotation Angle parameters to specify rotation values in X, Y, and Z. Incremental Rotate: This allows you to rotate the SPECT image in increments specified by the Rotation Increment control (and set as defaults: see Preferences for the ACMap Workstep on page 104). To rotate, click on the image. The image is divided in half by an imaginary vertical line. Clicking on the right half rotates the image clockwise; clicking on the left half rotates it counterclockwise. You can also use the Rotation Angle parameters to specify rotation values in X, Y, and Z. Reset: This puts the SPECT image in the position and orientation it was in when the workstep began. NM Application Suite Release 1.0 AutoSPECT Pro 75

85 3.1 The AC Map Workstep Tool Description Center: This centers the SPECT image on the CT image. If you have made any adjustments to the position or orientation of the SPECT image, they are overridden. Undo: This undoes the previous operation. Successive clicks undo each operation you performed, in order. Redo: If you have used Undo, this redoes the previous operation. Successive clicks redo each operation you performed, in order. Figure 35 Triangular quadrants used in incremental translation If you have a dataset acquired with the Philips fiducial markers (used as a QC tool for Precedence and BrightView XCT, for example), more controls become available: Figure 36 ACMap workstep controls with fiducial marker functions 1. Find Markers 2. Register 3. Hide Markers 4. Mean Distance Find Markers: This searches for markers and indicates their positions with a cross. Philips Healthcare Register: This registers the images based on markers that were found. Hide Markers: This toggles the display of the crosses. 76 AutoSPECT Pro NM Application Suite Release 1.0

86 The AC Map Workstep 3.1 Mean Distance: This is the average of the distance between all the fiducial markers in the SPECT image and the corresponding markers in the CT image for all markers that were found Using Vantage AC Data If you are using Vantage AC data, you can perform transmission reconstruction using some of the same controls available in the Reconstruction workstep, but with different options. Method (FBP) This is a standard filtered back project reconstruction. When this is selected, you must also set the following controls: Control Filters Values Butterworth (Off, Smoothing, Analytic) Cutoff Order Bound On or Off Philips Healthcare Rev A Filter Filter defaults to Butterworth. This is a low pass filter that smooths an image. You can control the smoothness by modifying the cutoff and order. It also has properties to ensure that the filter is applied consistently to files acquired with 64 x 64, 128 x 128, and 256 x 256 matrices. Additionally, you can determine the smoothing applied to the reconstructed transverse dataset: Off No smoothing is applied to the reconstructed or projection dataset. Smoothing A 3 x 3 spatial filter is applied to each slice of the reconstructed dataset. Analytic A filter based on the settings in the Filter, Cutoff, and Order fields is applied to the reconstructed dataset. The Analytic filter is a three-dimensional filter that smooths the reconstructed dataset in the x-, y-, and z-axis. Filtering along the x-axis and y-axis smooths the counts within each slice. Filtering along the z-axis smooths the counts from slice to slice. NM Application Suite Release 1.0 AutoSPECT Pro 77

87 3.1 The AC Map Workstep Cutoff Order Bound This reduces or eliminates the effects of high frequency information by applying a frequency cutoff value. When using filters to enhance nuclear medicine images, organs or areas of uniform counts are converted to low frequency signals, and lesions with sharp edges or background noise are converted to high frequency signals. Reducing the cutoff value smooths the image by eliminating the high frequency signals. Increasing the cutoff value sharpens the image by retaining the high frequency signals. However, both the lower and higher frequencies are attenuated. Drag the slider or type a value to set the cutoff value (the range is 0.1 to 2.00). IMPORTANT: The Cutoff and Order fields can be modified when a filter uses these fields, but they are grayed out when they are not in use by the filter. This modifies the exponent that determines the rate that a filter attenuates a signal. The filter order modifies the region where a filter goes from passing information to attenuating information. Decreasing the order widens the transition band, that decreases the attenuation rate of high frequency signals. Increasing the order narrows the transition band, that increases the attenuation rate of high frequency signals. The range is 0.0 to To increase or decrease the Order value, enter a value in the field or drag the slider. Check this to determine the body contour and reconstruct the data within that contour. Uncheck it to reconstruct all of the data within the field-of-view. Method (Bayesian) Control The Bayesian method uses a gradient iterative reconstruction method that assumes that the values within soft tissue will have a uniform attenuation coefficient. The output of this algorithm is significantly smoother and more accurate than the FBP result. This is the recommended method for performing transmission reconstruction. When Bayesian is selected, you must also set the following controls: Iterations 12 Values Philips Healthcare Start Truncation Correction Uniform On or Off 78 AutoSPECT Pro NM Application Suite Release 1.0

88 The Reconstruction Workstep 3.2 Iterations Start This control is only enabled if the selected method is Bayesian Iterative. The number of iterations depends on your data and preferences. Philips recommends 12 iterations. The initial estimate used by the iterative algorithm can be either a FBP image or an image with a uniform pixel value. FBP requires fewer iterations, and is recommended. Uniform starts with equalized pixel values and requires approximately five additional iterations to converge. Truncation Correction This is only available when the Bayesian Iterative method is selected. Enabling this applies a symmetry prior to correcting for truncation present when the patient is larger than the camera field of view. This method is identical to the regular BITGA method when no truncation is present. (BITGA, or Bayesian Iterative Transmission Gradient Algorithm, uses a prior function that preferentially weights the current attenuation coefficient estimate at each pixel toward the value for soft-tissue region.) The only reason not to use truncation correction is if there are artifacts present in the reconstructed transmission image Using Chang s AC If you are using Chang s AC, you can draw an ROI and then use these controls: Philips Healthcare Rev A Identically in all Frames: This uses the ROI in the same size and position in all frames. Individually Frame by Frame: This adjusts the ROI to the frame contents on every frame. Isotope: This dropdown menu allows you to select the isotope used in the study. Coefficient: This allows you to type in an attenuation coefficient value. Reset: This resets the ROIs to their state when the workstep began. 3.2 The Reconstruction Workstep The Reconstruction workstep (Figure 37) contains one or two sets (depending on the data loaded) of the following: NM Application Suite Release 1.0 AutoSPECT Pro 79

89 3.2 The Reconstruction Workstep Viewports that display projection data with or without motion correction, filtered projection data, and a reconstructed slice. Zoom controls that allow you to specify the zoom factor applied. Method controls that determine the reconstruction method used and any corrections to apply. Correction controls to apply to the data. Filter controls that determine the filter settings used to reconstruct the selected datasets. Limit parameters to specify the reconstruction range. The reconstruction methods, controls, and suggested parameters depend on the dataset type selected for processing. However, some of the controls and recommended settings are the same for all datasets. Figure 37 A Reconstruction layout To use the Reconstruction page, first adjust the start and end slice of the dataset. For cardiac datasets, these are set by the software, but you can adjust them to improve the results. You can either drag the reconstruction limit lines with the mouse or type values into the Start and End fields. After adjusting the slices, use the controls described in the following sections. Philips Healthcare 80 AutoSPECT Pro NM Application Suite Release 1.0

90 The Reconstruction Workstep 3.2 CAUTION: Visually verify the reconstruction results before saving to ensure images are reconstructed properly. Improperly reconstructed images may result in misdiagnosis Start/End Start and End display the start slice and end slice of the reconstructed transverse dataset. You can change the values by dragging the reconstruction limit lines or by typing values into the fields. NOTE: If you have multiple reconstructions of the same data, adjusting the slice limits for one reconstruction setting automatically adjusts them for all other settings. Similarly, if you have concurrent data (for example, gated and summed, dual isotope, etc.), adjusting the slice limits for one automatically adjusts them for the other Zoom Zoom allows you to apply a zoom factor to the reconstructed datasets. AutoSPECT Pro applies the zoom factor to the reconstructed transverse dataset. Short axis, horizontal long axis, and vertical long axis datasets created from this transverse dataset are reconstructed using the same zoom factor. NOTE: Zoom is disabled for all methods except FBP. Checking Zoom displays a box on the reconstructed slice. The box shows the image area corresponding to the zoom value. You can drag the box to reposition it on the image. Drag the zoom slider or type a value to set the zoom factor (the range is 1.0 to 3.0) Method AutoSPECT Pro provides four methods for reconstructing the transverse data. Philips Healthcare Rev A Standard FBP - Filtered Backprojection: Reconstruction is performed by backprojecting with a ramp filter. Iterative MLEM - Maximum Likelihood Expectation Maximization: Reconstruction is performed in an iterative fashion, with updates based upon a comparison of the estimation to the measured projection data. This method reduces streak artifacts found in FBP reconstructions. NOTE: MLEM only supports attenuation correction for Tc-99m and Tl-201. To use AC for other data, use OSEM or Astonish instead. 3D OSEM - Ordered Subsets Expectation Maximization: Reconstruction is similar to MLEM method, but only a subset of the projections NM Application Suite Release 1.0 AutoSPECT Pro 81

91 3.2 The Reconstruction Workstep are used in each iteration. This method offers faster reconstruction than MLEM with similar accuracy. Astonish - If you have the Astonish option enabled, you can select this method. Astonish uses Ordered Subsets Expectation Maximization with resolution recovery to reconstruct the dataset. Resolution recovery compensates for detector and collimator performance as a function of radius from the detector to the center of the image. It uses a deconvolution method based on measured collimator parameters that are stored in a configuration file. It also includes noise control that is independent of the number of iterations and subsets used. For more information on Astonish, see the Astonish Reconstruction on page 163 appendix. NOTE: Vantage data that is labeled _EMSCR has already had resolution recovery applied. It is inappropriate to apply Astonish to this data, so it is not available for _EMSCR data. After changing the method, review the available controls to be sure they are set as you intended. Refer to the table below for specific information about the controls. The controls available for each method are listed in the table below. Control FBP MLEM OSEM Astonish Bound On or Off Start FBP or Uniform Uniform Uniform Iterations 12 is recommended - then evaluate the image (range is 1-60) 2 is recommended - then evaluate the image (range is 1-60) 4 is recommended - then evaluate the image (range is 1-60) Subsets Filter None, Butterworth, Gaussian, Hamming, Hanning, Parzen Butterworth Butterworth Hanning, None Y-Axis Off, Smoothing, Analytic, Prefilter Off, Smoothing, Analytic Analytic Cutoff Order Start and End First and Last Slice Numbers First and Last Slice Numbers First and Last Slice Numbers First and Last Slice Numbers Philips Healthcare Matrix Same as input matrix, 64, 128, 256 Zoom AutoSPECT Pro NM Application Suite Release 1.0

92 The Reconstruction Workstep 3.2 Control FBP MLEM OSEM Astonish Attenuation Correction On or Off On or Off On or Off Scatter Correction On or Off On or Off Decay Correction On or Off On or Off On or Off On or Off Number of Detectors Single, Dual, Triple (auto-selected based on image header) Single, Dual, Triple (auto-selected based on image header) Single, Dual, Triple (auto-selected based on image header) Single, Dual, Triple (auto-selected based on image header) Axis Correction Motion Correction Motion Correction method Changs AC On or Off On or Off On or Off On or Off Auto, Manual Auto, Manual Auto, Manual Auto, Manual On or Off Bound This control is only enabled if the selected method is Iterative MLEM. Checking this determines the body contour and reconstructs the data within that contour. If it is unchecked, all of the data within the field-of-view is reconstructed Iterations Philips Healthcare Rev A Matrix Size This control is only enabled if the selected method is Iterative MLEM, 3D OSEM, or Astonish. The number of iterations depends on your data and preferences. The number of iterations is also affected by the initial estimation used. If FBP is used, 10 to 15 iterations are usually adequate. The effect of this control on an image involves a tradeoff. Generally, higher values yield a sharper image, but at the expense of increased noise. The range is The Matrix Size menu is only available when the method is FBP. Matrix allows you to specify the matrix size of the saved reconstructed datasets. The X1 option creates a reconstructed dataset that contains the same matrix size as the input dataset. Selecting 64, 128, or 256 creates a saved reconstructed dataset that contains the specified matrix size. NM Application Suite Release 1.0 AutoSPECT Pro 83

93 3.2 The Reconstruction Workstep IMPORTANT: Reconstructed datasets using a 256 matrix sizes can become very large, and may be incompatible with some software (AutoQuant, for example) Subsets This control is only enabled if the selected method is 3D OSEM or Astonish. The number of subsets depends on your preferences, and the number of projections in the loaded dataset. The number of projections divided by the number of subsets should be a whole number. For instance, if there are 64 projections, there could be 64, 32, 16, 8, 4 or 2 subsets. For 120 projections, there could be 120, 60, 40, 30, 15, 10, 5, 3, or 2 subsets. Requesting more subsets does not extend the reconstruction time, but requesting more iterations does. If you enter an invalid number of subsets, your value is replaced by the nearest valid value. Using the full number of projections as the number of subsets is not recommended. Philips recommends using a subsets value that is close to the number of projections divided by 4. Using 1 subset is valid, but this is essentially the same as using Iterative MLEM reconstruction, and does not take advantage of the ordered subsets algorithm Start This control is only enabled if the selected method is Iterative MLEM (it is set to Uniform for OSEM and Astonish). The initial estimate used by the iterative algorithm can be either a FBP image or an image with a uniform pixel value. FBP requires fewer iterations, and is recommended. Uniform starts with equalized pixel values and requires approximately five additional iterations to converge Attenuation Correction If you are processing a Vantage or CT-AC dataset or using a Changs AC preference, check this to apply attenuation correction to the reconstructed dataset Scatter Correction IMPORTANT: Do not neglect to check this option for Vantage datasets, or attenuation correction will not be applied. Philips Healthcare If you are processing a CT-AC or Vantage dataset, or using a saved attenuation map with OSEM or Astonish as the reconstruction method, you can check this to apply scatter correction during reconstruction. Scatter correction is based upon an estimated slab model. See References 7 84 AutoSPECT Pro NM Application Suite Release 1.0

94 The Reconstruction Workstep 3.2 and 9 in the NM Application Suite Reference Manual for a full description and validation of the method. Scatter correction can reduce the impact of detecting scattered photons from hot organs near the organ of interest. Scatter correction is only available when OSEM or Astonish is the reconstruction method, and when Attenuation Correction is ON. In general, Philips recommends using scatter correction whenever you perform OSEM or Astonish with Attenuation Correction. If you want to do Attenuation Correction only, without scatter correction, Philips recommends using Iterative MLEM with Attenuation Correction. NOTE: Vantage data that is labeled -EMSCR has already had scatter correction performed. It is inappropriate to apply AutoSPECT scatter correction to this data, so the scatter control is disabled if -EMSCR data is loaded. Scatter Correction uses a pre-calculated estimation (a kernel) of the scatter expected from a point source in water, which is stored on your hard drive. These kernels are specific to the isotope used, the energy window, and the pixel size. You can only perform scatter correction for combinations of parameters for which a kernel exists. The current release of AutoSPECT Pro supports scatter correction for seven isotopes; the following combinations of parameters are supported: Isotope Window 1 settings Window 2 settings Window 3 settings Supported Zoom Factors Tc-99m 15 or 140 kev NONE NONE All zooms Tl or 72 kev 15 or 167 kev (or NONE) NONE All zooms In or 173 kev 15 or 245 kev NONE All zooms Ga or 92 kev 15 or 185 kev 15 or 300 kev (or NONE) All zooms Philips Healthcare Rev A I or 159 kev I or 364 kev Lu or 113 kev NONE NONE All zooms NONE NONE All zooms 15 or 208 kev (or NONE) NONE All zooms NOTE: Using scatter correction in AutoSPECT may add several minutes to your reconstruction times, especially for isotopes with multiple scatter windows (e.g., Ga-67). NM Application Suite Release 1.0 AutoSPECT Pro 85

95 3.2 The Reconstruction Workstep Decay Correction Check this to apply decay correction to the projection dataset prior to reconstruction. This automatically pulls isotope information from the projection dataset. The Detector is automatically determined based on the dataset s header information. However, if you need to change it, select Single, Dual, or Triple, depending on how the dataset was acquired. IMPORTANT: Because of DICOM issues on some systems, the number of detectors may be misidentified. Be sure to always check this field and change it if necessary Motion Correction If Motion is checked in a Preference and it is set to Automatic, motion correction will happen with no intervention. However, if AutoProceed is off for Reconstruction, automatic motion correction happens when you enter the Reconstruction workstep, so you see the corrected images when the application pauses at the Reconstruction workstep. Also, if Motion is set to Manual but AutoProceed is on, Manual will override the autoproceed and the study will pause at the Reconstruction workstep in the motion correction layout. If you are in the Reconstruction workstep you can still perform automatic motion correction. To do this, Motion must be checked and set to Automatic. As soon as both of these conditions are met, motion correction is performed. There are two ways to perform and review motion correction: Automatic Motion Correction: Use this for Cardiac datasets to have the AutoSPECT Pro algorithm automatically locate the heart, evaluate it for motion artifacts, and make the appropriate adjustments. Selecting this automatically selects Auto as the mode, and performs the automatic corrections. The motion corrected projection image will cine in the upper viewport. Review the images in cine mode to determine if motion correction has been successfully applied. The heart should remain at the same level for all projections no sudden jumps should occur. Manual Motion Correction: Use this for non-cardiac datasets, and for cardiac datasets where the automatic motion correction induces artifacts. To perform manual motion correction, see the section below on Using Manual Motion Correction on page 90. If a Gated dataset is selected, any correction made to a Summed file is automatically made to the corresponding Unsummed file. Philips Healthcare 86 AutoSPECT Pro NM Application Suite Release 1.0

96 The Reconstruction Workstep 3.2 NOTE: If you check Motion after making manual corrections, AutoSPECT Pro replaces the manual corrections with automatic corrections. To use automatic motion correction, the Motion box must be checked in addition to Automatic being selected Axis Correction Axis applies an axis of rotation correction or center of rotation (COR) to the reconstructed datasets. Check Axis to enable the option. Then enter the amount of correction. This correction is only necessary for older Philips/ADAC cameras. It is not necessary if the camera includes this information in the acquired data as the newer Philips cameras do: Skylight, Forte, BrightView, and Precedence, for example. NOTE: This is only available when the selected Method is Standard FBP Filter Filter allows you to select the filter applied to the data during reconstruction. When the reconstruction method is FBP, each filter modifies a ramp filter to smooth or enhance image details. When the reconstruction method is MLEM or OSEM, the filter is applied after reconstruction. When the reconstruction method is Astonish, the filter is applied to the projections before and during reconstruction. Depending on the reconstruction method, the Filter menu contains combinations of the following options: None No filter modifies the ramp filter applied to the data during reconstruction. This may produce sharp but noisy images. Philips Healthcare Rev A Butterworth Gaussian This filter is available only with the FBP, MLEM, and OSEM methods, not Astonish. The Butterworth filter is a low pass filter that smooths an image. You can control the smoothness by modifying the cutoff and order. It also has properties to ensure that the filter is applied consistently to files acquired with 64 x 64, 128 x 128, and 256 x 256 matrices. This filter is available only with the FBP method. The Gaussian filter is a frequency filter based on an exponential function that takes the form: F(x) = a*exp(-b), where a and b are based on the mean and standard deviation. This equation is referred to in statistics as normal or bell curve. Because of its exponential drop-off, it also behaves well in filtering. NM Application Suite Release 1.0 AutoSPECT Pro 87

97 3.2 The Reconstruction Workstep An advantage of the Gaussian filter is that it can become almost any kind of filter (low pass, high pass, or band pass) and it can have a gradual or sharp cutoff. The shape of the filter is controlled by the cutoff frequency and the filter order. It also has properties to ensure that the filter is applied consistently to files acquired with 64 x 64, 128 x 128, and 256 x 256 matrices. The cutoff frequency is entered as a percentage of the Nyquist frequency. The Nyquist frequency is the highest possible frequency in the image. The resolution limit is set by the linear sampling distance. Therefore, the cutoff frequency can range from 0.0 to 1.0. A cutoff frequency of 0.5 represents 50% or 1/2 of the Nyquist frequency. The filter order specifies the steepness of the cutoff. In general, the higher the filter order, the steeper the cutoff. A steep cutoff is necessary to eliminate structures that may overlap in frequency. For example, if you have high frequency data and high frequency noise, a sharp cutoff helps isolate the data. Otherwise, a gradual cutoff is more desirable. Choice of filter parameters is dependent on the nature of the data, personal preference of the user, counting statistics and camera performance. Hamming Hanning Parzen This filter is available only with the FBP method. It is a modified Hanning window with a more abrupt cutoff at the high frequency limit. This filter is available only with the FBP and Astonish methods. It is a low pass filter that smooths an image. Its falloff, which controls the attenuation of the high frequencies, is determined by the cos 2. This filter is available only with the FBP method. It is a low-pass filter that smoothes an image using a linear falloff Y Axis This menu is unlabeled, but is next to the Filters menu. Off Smoothing No smoothing is applied to the reconstructed or projection dataset. This filter is available only with the FBP, MLEM methods. A 3 x 3 spatial filter is applied to each slice of the reconstructed dataset. Philips Healthcare 88 AutoSPECT Pro NM Application Suite Release 1.0

98 The Reconstruction Workstep 3.2 Analytic Prefilter This filter is available only with the FBP, MLEM, and OSEM methods. A filter based on the settings in the Filter, Cutoff, and Order fields is applied to the reconstructed dataset. The Analytic filter is a three-dimensional filter that smooths the reconstructed dataset in the x-, y-, and z-axis. Filtering along the x-axis and y-axis smooths the counts within each slice. Filtering along the z-axis smooths the counts from slice to slice. This is a smoothing filter used before reconstruction. This filter is used only with the Standard FBP method Cutoff This reduces or eliminates the effects of high frequency information by applying a frequency cutoff value. When using filters to enhance nuclear medicine images, organs or areas of uniform counts are converted to low frequency signals, and lesions with sharp edges or background noise are converted to high frequency signals. Reducing the cutoff value smooths the image by eliminating the high frequency signals. Increasing the cutoff value sharpens the image by retaining the high frequency signals. However, both the lower and higher frequencies are attenuated. Generally, too low a cutoff value produces over-smoothed data, possibly disguising a lesion. Too high a cutoff value produces noisy images which appear patchy. The filter choice must reflect both the frequency context of the noise and the frequency context of the organ in the image. Drag the slider or type a value to set the cutoff value (the range is 0.1 to 2.00). IMPORTANT: The Cutoff and Order fields can be modified when a filter uses these fields, but they are grayed out when they are not in use by the filter. Philips Healthcare Rev A Order The Order value, used in the Butterworth and Gaussian filter functions, modifies the exponent that determines the rate that a filter attenuates a signal. The filter order modifies the transitional band (the region where a filter goes from passing information to attenuating information). Decreasing the order widens the transition band, which decreases the attenuation rate of high frequency signals, and can decrease the smoothing. Increasing the order narrows the transition band, which increases the attenuation rate of high frequency signals, and can increase the smoothing. So the higher the filter order, the narrower the transition band and the steeper the drop-off. NM Application Suite Release 1.0 AutoSPECT Pro 89

99 3.2 The Reconstruction Workstep The value ranges from 0.0 to To increase or decrease the Order value, enter a value in the field or drag the slider Comparing Parameters You can compare parameter settings using the Compare page. This displays three copies of a slice, and all the controls available for the data. The parameter values on the Reconstruction page are used for the middle slice; for comparison, the top slice has slightly lower values, and the bottom slice has slightly higher values. You can adjust any of the values to create a more meaningful comparison. When you are done adjusting the copies, click Apply for the one that you want to save. For studies with multiple datasets, click Apply once for each dataset. Click the Reconstruction button to go back to the Reconstruction page. The reconstructed slice will reflect the changes you made, and the controls will retain the settings you made Using Manual Motion Correction Use the Motion Correction feature to analyze and correct for patient motion that may have occurred during an acquisition or to verify that motion artifacts have been corrected when automated motion correction is applied. Perform motion correction on summed emission data only. This feature is not available for transmission data, and should be used for review purposes only for unsummed (gated) data. Unsummed data is automatically summed when loaded, and you can motion correct that. The motion corrected values are automatically applied to the unsummed data as well, and view it using a layout for the gated data. The Motion Correction panel displays the original and corrected projection dataset with sliders and horizontal reference bars, a sinogram or cyclogram of the original projection and corrected dataset, a motion graph, and the current slice: Philips Healthcare 90 AutoSPECT Pro NM Application Suite Release 1.0

100 The Reconstruction Workstep 3.2 Figure 38 Motion correction page Philips Healthcare Rev A 1. Horizontal reference bars 2. Motion correction sliders 3. Motion graph of slices Here are the controls for Motion Correction: NM Application Suite Release 1.0 AutoSPECT Pro 91

101 3.2 The Reconstruction Workstep Figure 39 Motion correction controls 1. Motion correction arrows 2. Reset 3. Show/Hide Markers 4. Enhance 5. Sinogram/Cyclogram 6. Mask See the next section for information on sinograms and cyclograms. Use the controls below to perform motion correction. CAUTION: Review the motion corrected projection datasets in cine mode to ensure that motion has been corrected for accuracy. Improperly applied motion correction may create artifacts resulting in misdiagnosis. Correct X, Y This allows you to apply automatic motion correction changes to the X and Y axes individually. By default, only Correct Y is checked. Show/Hide Markers This toggles the display of the hash marks and coordinates in the Corrected viewport. Enhance Sinogram/Cyclogram Use this to display the pixel count differences between frames. Image brightness is proportional to the amount of inter-projection movement present. When displayed with cine on, the areas of motion artifacts are accentuated. By default, a sinogram is displayed when the Motion Correction page appears. A sinogram assists in identifying vertical motion, whereas a cyclogram assists in identifying horizontal motion. You can toggle between a sinogram and a cyclogram using this button (this control is also available Philips Healthcare 92 AutoSPECT Pro NM Application Suite Release 1.0

102 The Reconstruction Workstep 3.2 when you right-click on a sinogram image). They both update as you make adjustments to an image. For more on sinograms and cyclograms, see the next section, Analyzing Sinograms and Cyclograms on page 95. Mask This allows you to specify the region of interest in a dataset (this button is also available when you right-click on a mask image). Drag the crosshairs in the center to move the mask; drag the small circles to resize the mask. 1. Move 2. Resize Figure 40 Mask controls Philips Healthcare Rev A Detector Manually Correcting for Motion Select the detector type (Single, Dual, Triple) used during acquisition from the Detector pulldown menu. See the note in the next section for information about applying manual motion correction to Dual-Head data. NOTE: This is automatically detected when the information is available in the patient header. However, be sure to review this setting to verify that it matches the acquisition, and change it if necessary. In all cases, you can use Single to correct any data, even if it was acquired with two or three heads. The program can use the information about the correct number of heads to speed up and simplify motion correction, but it can successfully motion correct each frame individually if Single is selected. To manually correct for motion, you need to step through each slice in the reference viewport, move the image up or down or side to side as needed to align the images, and reconstruct the image for motion correction. NM Application Suite Release 1.0 AutoSPECT Pro 93

103 3.2 The Reconstruction Workstep IMPORTANT: To minimize the number of slices you need to correct, examine the entire dataset for motion and note the slices where correction is needed. Determine the minimum number of slices you can correct to align the entire study. For example, if slices one and two are offset from all other slices, correct the first two slices relative to the last. NOTE: When applying manual motion correction to dual-head data, the software enforces the following rules: For Dual- and Triple-Head datasets, corrections in the Y direction that are applied to the Nth projection image of one detector head are also applied to the Nth projection image of the other detector head. For example, if you correct frame 6 of a 64 frame dual head study by an amount 0.5 pixels, NM Application Suite will also correct frame 38 (6 + 32) by an amount 0.5 pixels. For Dual-Head datasets that are acquired in the relative 180 degree configuration, an additional rule is enforced (but not for the 90 degree configuration). Corrections in the X direction that are applied to the Nth projection image of one detector are applied in the opposite direction to the Nth projection of the other detector. For example, if you correct frame 6 of a 64 frame dual head study by an amount 0.5, Autospect will also correct frame 38 (6 + 32) by an amount -0.5 pixels. For Triple-Head data, corrections in the X direction on one head must take into account the angles of the other heads when calculating corrections for them. This means that, for example, correcting by 0.5 pixels on one head may mean correcting by 0.25 on another, and by on the other. For the following procedure, refer to Figure 38 above. To manually correct a dataset for motion: 1. Analyze the sinogram, cyclogram, or the cine display of the projection dataset to determine if the study contains a motion artifact. To accentuate motion artifacts, click on Enhance and analyze the cine display of the projection dataset. With Enhance enabled, the pixel count difference between frames is displayed. Therefore, regions of increased motion are displayed as brighter regions. 2. In the Original Data viewport (labelled at the bottom of the viewport), align the three horizontal reference bars with a constant point of reference. This automatically adjusts the reference bars in the Motion Corrected Data viewport. For example, in a cardiac study, align the top reference bar with the superior ventricular surface, the bottom reference bar with the inferior ventricular surface and the middle reference bar with the mid-ventricular cavity. You can use the reference lines as a visual cue when determining whether the patient has moved. It may help to Ctrl-click the viewport and use the cine controls in the Viewer Tools manager. Philips Healthcare 94 AutoSPECT Pro NM Application Suite Release 1.0

104 The Reconstruction Workstep Step through the projections using the arrow keys on the keyboard. You can also use the mouse wheel to step through the projections. Analyze the image as you step through the projections until you reach the image needing correction. 4. Align the image in the Corrected viewport with the reference bars, as necessary. To do this, do one of the following: Click on the motion correction arrows (see the figure above) to move the image vertically or horizontally 0.1 pixel per click. NOTE: Click the RESET button (see the figure above) to return to the original settings of the currently active frame. Drag the motion correction sliders (the yellow hash marks) in the Corrected viewport to move the image vertically or horizontally. NOTE: The pixel values appear in the upper right corner of the Corrected viewport. As you move the images a bar appears in the Graph viewport (see the figure above). These bars give a visual representation of where the slices are motion-corrected, how far they have been corrected, and in what direction. IMPORTANT: Do not press Correct when manually correcting for motion. If you press Correct after manually moving the data, your entries will be overwritten by the automatic motion control. When the image is reconstructed the motion correction changes are applied Analyzing Sinograms and Cyclograms AutoSPECT Pro allows you to create sinograms and cyclograms to analyze motion artifacts in SPECT studies. Philips Healthcare Rev A Understanding Sinograms To create a sinogram, a single row of pixel values in each projection image forms a row of pixel values in the sinogram (Figure 41). For example, to create a sinogram displaying the data used to create the Nth slice in a reconstructed dataset, the count values contained in the Nth row of each projection image would be assigned to the corresponding row in the sinogram. The count values contained in the Nth row of Image 1 form the first row of the sinogram. The count values contained in the Nth row of Image 2 form the second row of the sinogram. The number of rows in the sinogram equals the number of projections in the SPECT dataset. The number of images in the sinogram dataset equals the number of slices in the reconstructed dataset. NM Application Suite Release 1.0 AutoSPECT Pro 95

105 3.2 The Reconstruction Workstep Figure Image 1 2. Image 2 3. Image 3 4. Nth slice 5. Row N from Image 1 6. Row N from Image 2 7. Row N from Image 3 Sinogram generation Analyzing Sinograms 1. If the cyclogram is displayed, click on Toggle Cyclogram in the Motion Correction window to switch from the cyclogram display to the sinogram display. 2. Display the sinogram from the desired slice. 3. Drag the reference line in the Reference viewport. The sinogram display is updated to the current slice. The viewport for the original SPECT dataset displays the dataset as a series of individual projection images. Each projection image contains the counts acquired by a detector at a specific azimuth. Displaying other projection images in the SPECT dataset allows you to view the counts acquired at different azimuths. However, an image in a sinogram displays the counts from the same row in all of the projection images that are used to create a single tomographic slice. Moving the reference line up or down displays sinograms at different slice levels. 4. Analyze the sinogram for evidence of motion. Viewing a dataset as a sinogram allows you to easily detect motion artifacts. A sinogram created from a dataset without motion artifacts appears as a Philips Healthcare 96 AutoSPECT Pro NM Application Suite Release 1.0

106 The Reconstruction Workstep 3.2 smooth spiral, but a sinogram created from a dataset containing motion artifacts contains horizontal breaks or discontinuities in the spiral (Figure 42). Figure 42 Sinogram without (left) and with (right) motion artifacts Understanding Cyclograms Cyclograms are similar to sinograms but are generated by selecting a point in a transverse slice and then concatenating from each projection image the vertical strip that passes through that point (Figure 43). Philips Healthcare Rev A 1. Image 1 2. Image 2 3. Image 3 4. Column N from Image 1 Figure 43 Cyclogram generation NM Application Suite Release 1.0 AutoSPECT Pro 97

107 3.3 The Reorientation Workstep 5. Column N from Image 2 6. Column N from Image 3 Analyzing Cyclograms 1. In the Motion Correction window, click on Cyclogram to enable the cyclogram display. A transverse sample slice will appear based on the reference line in the original projection image. 2. Drag the reference line in the original viewport. The sample transverse slice is updated to the current slice corresponding to the position of the reference line. 3. Use the crosshair in the sample transverse slice to determine the transverse column to be used to generate the cyclogram. Ensure that the crosshair is within the correct ROI. If the crosshair is not within the ROI, right-click and drag the crosshair within the ROI. 4. Analyze the cyclogram for evidence of motion. Cyclograms display evidence of motion similar to sinograms except in the vertical plane. A cyclogram created from a dataset without motion artifacts appears as a smooth wavy image, but cyclograms created from datasets containing motion artifacts contain vertical breaks or discontinuities in the image (Figure 44). Figure The Reorientation Workstep Cyclogram without (left) and with (right) motion artifacts Philips Healthcare The Reorientation workstep (Figure 46) allows you to automatically or manually reorient SPECT or gated SPECT datasets. When you select the Reorient workstep, cardiac datasets are automatically processed and 98 AutoSPECT Pro NM Application Suite Release 1.0

108 The Reorientation Workstep 3.3 reoriented. If you are not satisfied with the automatically identified reorientation, or if the dataset is non-cardiac, you must adjust the limits manually. NOTE: Reoriented data is saved in the ACC format: the apex points to the right in the vertical long axis, and upward in the horizontal long axis (see the figure below). Additionally: Short Axis images in the dataset are sliced from apex to base. Vertical Long Axis images in the dataset are displayed from septum to lateral wall. Horizontal Long Axis images in the dataset are displayed from inferior to superior. Figure 45 Cardiac ACC orientation (SAX at left, VLA in middle, HLA at right): Philips Healthcare Rev A 1. Superior 2. Inferior 3. Septal 4. Lateral 5. Base 6. Apex Perform reorientation on summed emission data only. Changes made to the reorientation on the unsummed gated data are applied to the Summed data, and vice-versa. Philips recommends making changes to the reorientation only on the Summed Emission data page. For multiple reconstructions of the same dataset in the same processing session (for example, with and without attenuation correction), changes made to the reorientation of one reconstruction are automatically made to any other reconstruction in the session. NOTE: If you have used scatter correction or Astonish for reconstruction, the transverse slice in the Reorientation page may look different from the sample Reconstructed Slice image on the Reconstruction page. This is because calculations for the sample slice do not use the full 2D or 3D information used for the final reconstruction displayed in the Reorientation page. NM Application Suite Release 1.0 AutoSPECT Pro 99

109 3.3 The Reorientation Workstep Figure 46 Reorientation page CAUTION: Visually verify the reorientation results before saving to ensure images are reconstructed properly. Improperly reoriented images may result in misdiagnosis Reorienting Datasets The Reorient workstep (Figure 46) contains Azimuth and Elevation reference lines that you can drag and rotate. If the Twist option is enabled, an axial tilt reference line is displayed in the bottom viewport. The orientation of the reference lines determines the orientation of the reconstructed datasets. Philips Healthcare 100 AutoSPECT Pro NM Application Suite Release 1.0

110 Setting Preferences for AutoSPECT Pro 3.4 To manually reorient the organ of interest: 1. Position each reference arrow in the organ of interest: for each viewport, drag the circle in the middle of the reference arrow to the center of the organ of interest. 2. Drag an end of a reference arrow to rotate it to align the organ of interest. Alternatively, type a value in the Azimuth and Elevation boxes at the bottom. 3. If necessary, check the Twist box at the bottom to display the Twist reference arrow and adjust that as well. 4. Specify the slices to save by dragging the reference lines on the right side of each viewport to enclose the organ of interest. Use the Matrix Zoom control at the bottom to control the apparent size of the image: either drag the slider or type in a zoom value (the range is 1-3). If you need a specific pixel size, use the Pixel Size box to type in a value. NOTE: You cannot adjust the viewports individually with Zoom. IMPORTANT: When positioning the reference lines, allow sufficient space between the lines and the organ so that you do not clip it. You can save the following data and image types: Data Type Tomo/Emission Gated Tomo Gated ReconTomo (Emission) Recon Tomo (Emission) Recon Tomo (Transmission) Image Type Motion Corrected Tomo Gated Motion Corrected Tomo Summed Tomo, Motion Corrected summed Tomo, Motion Corrected gated Tomo, Gated Transverse, Gated Short Axis, Gated Horizontal Long Axis, and Gated Vertical Long Axis Non cardiac: Oblique/Reoriented Transverse, Sagittal, and Coronal Cardiac: Transverse, Short Axis, Horizontal Long Axis and Vertical Long Axis AC Map Philips Healthcare Rev A 3.4 Setting Preferences for AutoSPECT Pro AutoSPECT Pro handles preferences differently from the other applications in the NM Application Suite. This section describes how to set preferences in AutoSPECT Pro only, not in any other application. There are 21 factory preferences for AutoSPECT Pro: 1 Day 2 Day Dual Isotope NM Application Suite Release 1.0 AutoSPECT Pro 101

111 3.4 Setting Preferences for AutoSPECT Pro Thallium Astonish Cardiac Vantage Pro Astonish Vantage Pro Astonish Cardiac CT-AC Cardiac CT-AC Bone Astonish Bone TB SPECT General Dual CT-AC General Chang AC General CT-AC Brain Astonish Brain Brain Chang AC Brain CT-AC Cardiac Reorientation Reorientation To use Preferences, select the Preferences Data Manager. This manager contains a list of all the AutoSPECT Pro preferences, and some controls at the bottom: Figure 47 Preferences controls 1. Edit Preference 2. Delete Preference 3. Apply Preference 4. Show Factory To edit a preference, select the preference from the list and click Edit Preferences. See the sections below for details on editing preferences. Philips Healthcare To apply a preference, double-click it, or select it and click Apply Preference. 102 AutoSPECT Pro NM Application Suite Release 1.0

112 Setting Preferences for AutoSPECT Pro 3.4 To delete a preference, select the preference from the list and click Delete Preference. To toggle the display of the factory-installed preferences, check the Show Factory checkbox General Preferences In an AutoSPECT Pro preference, each workstep has its own preference settings. In addition, you can set some parameters for the preference as a whole (ones that apply to all worksteps): AutoProceed: This determines whether the workflow automatically continues through the workstep if all the required input is present. If this is unchecked, the workflow will always stop at the workstep. If it is checked, it will stop only if the required input is not present. NOTE: AutoProceed is available for all worksteps except the first one, Setup. You must manually proceed from the Setup step even if AutoProceed is checked for it in the Preference. Acquisition Matching String: This allows you to set strings for exam matching based on the DICOM attributes Study Description, Protocol Name, or Body Part Examined. Separate strings with a comma. Matching succeeds if any of the DICOM attributes contains one of the matching strings. CT Window Preset: These translate the values of an image into a range of gray levels suitable for optimal viewing of the specified organ or area. SPECT ColorMap: These convert gray levels to colors using different schemes. NOTE: CT images loaded into the AC Map workstep are always displayed in grayscale. All other parameters are specific to the worksteps and are described below. Philips Healthcare Rev A Preferences for the Setup Workstep The following controls appear in the Setup workstep. Matching String: Strings you type in here are used to match datasets and exams that will use the preference. Separate strings with a comma. Matching succeeds if the value of any of these DICOM attributes contains one of the matching strings: For datasets: Study Description, Protocol Name, and Body Part Examined (organ) NM Application Suite Release 1.0 AutoSPECT Pro 103

113 3.4 Setting Preferences for AutoSPECT Pro For exams, the matching proceeds in this order: Acquisition Context, Image ID, Series Description and Image Comment Active: Check this to set whether the bucket appears in the Setup step. Alternate Name: This is an additional string to use with the default bucket name Preferences for the ACMap Workstep The controls available in this workstep depend on the preference you are editing. For a Changs AC preference, there are no relevant controls. For a Vantage preference, the controls are a subset of the Reconstruction workstep controls (Method, Corrections, Filters, etc.); see the descriptions in the section on The Reconstruction Workstep on page 79 for details. For most other preferences, the following controls are present: Rotation: This sets the increment by which the SPECT image rotates when you click on it using the Incremental Rotate tool. Translation: This sets the increment by which the SPECT image moves when you click on it using the Incremental Translate tool Preferences for the Reconstruction Workstep The following controls appear in the Reconstruction workstep. Since you can have up to three reconstructions when using the Compare function, you can set the preferences for each one using the buttons on the left. Click a button to select it, and set the preferences described below. Use the checkboxes to specify whether the reconstruction is available. Auto Knitting: Checking this automatically knits loaded segments during reconstruction. However, for this to work, all segments must include the full field of view, and they must all use the same reconstruction settings (Method, Iterations, Filter, etc.). Mode: This determines whether there is a single set of reconstruction and reorientation controls or a dual set. Reconstruction controls: The Reconstruction controls allow you to set the default settings seen in the workstep: the default Method, Iterations, Filter, whether the Corrections are on or off, etc. For explanations of these, see the sections that describe them. NOTE: If Motion is checked and set to Manual, this will override the AutoProceed setting, and the process will stop at the Reconstruction workstep to allow manual motion correction. Philips Healthcare 104 AutoSPECT Pro NM Application Suite Release 1.0

114 Setting Preferences for AutoSPECT Pro 3.4 Saving Options: These are the datasets to save. Check a dataset to have it saved; type a string in the text box to save the dataset with that name instead of the one displayed. Buckets: This specifies the bucket that is to be a required dataset in the Setup workstep Preferences for the Reorientation Workstep The following controls appear in the Reconstruction workstep. Oblique Twist: This determines whether Twist is on or off. Matrix Zoom: This specifies a default zoom factor Preferences for the Review Workstep This is a list of all the viewing protocols. Select a Default protocol by clicking on its radio button. Select the protocols to make available by using the Active checkboxes. For protocols you have created (listed under User Viewing Protocols), you can check Delete Protocol, which deletes the protocol on saving the Preference Saving and Applying Preferences When you are done making changes to a preference, you have the following options: IMPORTANT: Before saving, be sure to verify that you have configured the options in all the appropriate worksteps. Philips Healthcare Rev A Save As: This displays a dialog that allows you to save the changes you have made as a new Preference. Save: This saves the changes you have made to the preference. To save it as a new preference, type a name in the Preference Name field before clicking Save. NOTE: If you make changes to a factory default Preference and Save it, a new user Preference is saved with the same name; the factory Preference remains unchanged. You can also use Save As and save it as a new Preference. NOTE: Save does not apply the changes to the current dataset; for that, use Save Apply (below). Apply: This applies the current settings to the workstep, but does not save them to the preference. Save Apply: This saves the changes you have made to the preference, and also applies it to the current dataset. NM Application Suite Release 1.0 AutoSPECT Pro 105

115 3.4 Setting Preferences for AutoSPECT Pro Close: This closes the Preferences Editor without saving changes AutoSPECT Pro Default Protocols Here are descriptions of all the default protocols used in AutoSPECT Pro. 1 Day 2 Day This protocol reconstructs gated and summed STRESS and REST cardiac studies without attenuation correction. This default uses FBP with parameters previously established in AutoSPECT Plus for a 1 day Cardiolite protocol. This protocol reconstructs gated and summed STRESS and REST cardiac studies without attenuation correction. This default uses FBP with parameters previously established in AutoSPECT Plus for a 2 day Cardiolite protocol. Astonish Cardiac CT-AC This protocol reconstructs gated and summed STRESS and REST cardiac studies with and without CT-based attenuation correction. This default uses Astonish for emission data with parameters suitable for cardiac studies. Astonish Cardiac This protocol reconstructs gated and summed STRESS and REST cardiac studies without attenuation correction. This default uses Astonish with parameters determined from clinical studies. See the NM Application Suite Reference Manual for references. Astonish Vantage Pro This protocol reconstructs gated and summed STRESS and REST cardiac Vantage studies with and without attenuation correction. This default uses Bayesian reconstruction for transmission data and Astonish for emission data with parameters determined from clinical studies. See the NM Application Suite Reference Manual for references. Bone SPECT Astonish This protocol reconstructs bone scans, or general SPECT studies, without attenuation correction. This default uses Astonish with parameters determined by clinical users. Philips Healthcare 106 AutoSPECT Pro NM Application Suite Release 1.0

116 Setting Preferences for AutoSPECT Pro 3.4 Bone SPECT Brain Astonish Brain Chang AC Brain CT-AC Brain Default Cardiac CT-AC This protocol reconstructs bone scans, or general SPECT studies, without attenuation correction. This default uses FBP with parameters previously established in AutoSPECT Plus. This protocol reconstructs brain studies without attenuation correction. This default uses Astonish with parameters previously established in AutoSPECT Plus. This protocol reconstructs brain studies with and without Chang's attenuation correction. This default uses FBP with brain parameters previously established in AutoSPECT Plus. This protocol reconstructs brain studies with and without CT-based attenuation correction. This default uses OSEM with generic parameters. This protocol reconstructs brain studies without attenuation correction. This default uses FBP with parameters previously established in AutoSPECT Plus. This protocol reconstructs gated and summed STRESS and REST cardiac studies with and without CT-based attenuation correction. This default uses OSEM for emission data with parameters previously established in AutoSPECT Plus. Philips Healthcare Rev A Cardiac Vantage Pro This protocol reconstructs gated and summed STRESS and REST cardiac Vantage studies with and without attenuation correction. This default uses Bayesian reconstruction for transmission data and MLEM for emission data with parameters previously established in AutoSPECT Plus. Cardiac Reorientation This protocol reorients previously reconstructed gated and/or ungated cardiac SPECT volumes. Dual Isotope This protocol reconstructs gated and summed STRESS and REST cardiac studies without attenuation correction. This default uses FBP with parameters previously established in AutoSPECT Plus for a thallium REST, technetium STRESS protocol. NM Application Suite Release 1.0 AutoSPECT Pro 107

117 3.4 Setting Preferences for AutoSPECT Pro General Chang AC General Dual CT-AC General CT-AC This protocol reconstructs general SPECT studies, both with and without Chang's attenuation correction. This default uses FBP with generic parameters. You can use this for concurrently reconstructing two SPECT data sets (e.g., ventilation and perfusion studies, dual isotope studies, etc.) both with and without CT-based attenuation correction. This default uses OSEM with generic parameters. This protocol reconstructs up to 3 TB SPECT segments concurrently both with and without CT attenuation correction. This default uses OSEM with generic parameters. General Reorientation This protocol reorients previously reconstructed non-cardiac SPECT volumes. TB SPECT Thallium This protocol reconstructs up to 6 TB SPECT segments concurrently. This default uses MLEM with no axial filter to minimize knitting artifacts, and no attenuation correction. This protocol reconstructs gated and summed STRESS and REST cardiac studies without attenuation correction. This default uses FBP with parameters previously established in AutoSPECT Plus for a thallium protocol. Philips Healthcare 108 AutoSPECT Pro NM Application Suite Release 1.0

118 4 Cardiac This application allows you to perform gated planar (MUGA) analysis, first pass analysis, and quantify left-to-right shunts. It has these methods: LV Muga GBP LV Muga C Shunt First Pass For information on loading requirements, and on calculations and algorithms used in this application, see the appropriate section in the NM Application Suite Reference Manual. 4.1 MUGA MUGA (Multi-Gated Acquisition) allows you to automatically segment and quantify gated blood pool datasets, and create statistical information about the cardiac cycle. A Left Ventricle ROI is required for MUGA, but the Right Ventricle ROI is also selectable as a Method. By default, bounding boxes are automatically drawn around the ventricles. From these regions the application determines ED and ES, and generates results using relevant corrections (for example, background correction) Using MUGA In the Define Regions workstep, this application draws a preliminary bounding box around one or more regions. By default the edge detection algorithm is GBP Multiple. You must then take these steps to create ROIs: Philips Healthcare Rev A 1. Move and adjust a bounding box so it encloses the appropriate ventricle. Drag the line to move the bounding box; drag a handle to reshape it. 2. Click Detect Region. This creates an ROI for the ventricle and an ROI for the background. To start over, click the circle icon again; to draw the ROI by hand, click the eraser icon. The ED and ES frames for LV and RV are automatically determined and the ROIs are automatically drawn. If you would like to change any of these: 1. In the Cine viewer, go to the correct frame (using the Scroll feature, for example). NM Application Suite Release 1.0 Cardiac 109

119 4.1 MUGA 2. If necessary, redraw the ROI by clicking on its eraser icon and using the ROI pencil tool. 3. Right-click in the Cine viewer, and use the Set As pop-up menu to select a label. In the Review Results workstep, you can use the red timing marker in the Time Activity viewer to get results values and Regional EFs at a specific time. Drag the marker to a specific time to view the values; the precise time appears just above the marker. CAUTION: If the count rate is significantly lower in the last few bins of the study, the Ejection Fraction value and the identification of the ED and ES frames can be wrong. To avoid this situation, use the timing marker in the Time Activity viewer to exclude the last bins from the calculations Results Cine ED/ES images Phase/Amplitude images, and Phase Histogram Ejection Matrix/Stroke images Ejection Fraction Background Counts Peak Filling Rate Peak Ejection Rate Time to Peak Filling Time to Peak Ejection Filling for 25%, 33%, 50%, and 75% Nominal Interval RR Interval Window Acquired Beats Rejected Beats Skipped Beats Volume Time Activity Curve Regional Ejection Fractions First Derivative Time Activity Curve Philips Healthcare 110 Cardiac NM Application Suite Release 1.0

120 First Pass 4.2 If you do not see all the result images in the Review Results workstep, it may be that one or more viewers are hidden. If you suspect this, try using the Show Hidden Viewers tool in the Utilities Data Manager. See the section on Review Results Workstep for details Preferences To change the Preferences for this application: 1. Select the Preferences Data Manager. 2. Click Open Preference Editor at the bottom of the Preferences section (the second icon: ). 3. Make changes in the preferences window using the information in the table below. See the section on Creating and Editing Preferences on page 43 for details on editing Preferences. You can save these parameters as Preferences: Parameter Default Description Muga Algorithms GBP Multiple This is the algorithm used for edge detection. View Smoothed Image True This determines whether the cine images are smoothed (using the Fourier algorithm) Slope 1.0 Slope of the line fit to points in a calibration plot Intercept 0.0 Y intercept of the line fit to points in a calibration plot 4.2 First Pass Philips Healthcare Rev A Using First Pass This calculates right or left ventricular ejection fractions from a First Pass dynamic study. It requires rapid framing rate studies (30-40 sec acquisitions of 0.03 to 0.04 secs per frame) of the first transit through the heart to obtain a 16-frame gated study. In the Define Regions workstep, after drawing the ROIs, use the timing markers in the Whole Heart Time Activity curve to specify the time to analyze. In the Review Results workstep, review the frame pairs in the Ventricle curve. These are the colored lines in the time activity curve. If they need redefining, follow these steps: NM Application Suite Release 1.0 Cardiac 111

121 4.2 First Pass 1. Select Manual from the Frame Pairs Identification Type pull-down menu. 2. Click on a FramePair radio button to select the pair to define. 3. Drag the red and green timing markers from their default position at time zero to define the beginning and end frames. 4. Click Add Frame Pair to confirm the settings. 5. Identify other frame pairs similarly. 6. Use the Delete Frame Pair button to delete a pair Results Cine display Gated Study display Ejection Fractions for each frame pair Average Ejection Fraction Ventricle Time Activity Curve Splash display If you do not see all the result images in the Review Results workstep, it may be that one or more viewers are hidden. If you suspect this, try using the Show Hidden Viewers tool in the Utilities Data Manager. See the section on Review Results Workstep for details Preferences To change the Preferences for this application: 1. Select the Preferences Data Manager. 2. Click Open Preference Editor at the bottom of the Preferences section (the second icon: ). 3. Make changes in the preferences window using the information in the table below. See the section on Creating and Editing Preferences on page 43 for details on editing Preferences. You can save these parameters as Preferences: Parameter Default Description Review Compress Factor 1 Number of frames to compress for review data Philips Healthcare No. of Frames 6 This determines how many frame pairs to identify in the Ventricle Time Activity curve in the Results step 112 Cardiac NM Application Suite Release 1.0

122 Shunt Shunt This calculates the pulmonary-to-systemic flow ratio to determine if a left to right cardiac shunt is present. Given a fast dynamic cardiac study (usually a 30 sec acquisition of 0.5 secs per frame), Shunt processing requires fitting of the pulmonary curve with a gamma variate fit, subtraction of the fit and subsequent fitting of the remaining curve data with a second gamma fit to determine the pulmonary-to-systemic flow ratio. Superior Vena Cava quality control time is also computed to evaluate the bolus quality Using Shunt In the Define Regions workstep, after defining the ROIs, use the timing markers in the Right Lung Time Activity curve to define the points to use to fit the gamma curve to the data as closely as possible. The Recirculation curve is drawn automatically, based on the Right Lung fitted curve. Use the timing markers in the Recirculation curve to fit its gamma curve as well Results Philips Healthcare Rev A Preferences Composite image Cine with all ROIs SVC Control Time Pulmonary Transit Time Qp/QS ratio SVC (superior vena cava) Time Activity curve Time Activity curves for right lung, gamma fit, recirculation, and recirculation fit Splash display If you do not see all the result images in the Review Results workstep, it may be that one or more viewers are hidden. If you suspect this, try using the Show Hidden Viewers tool in the Utilities Data Manager. See the section on Review Results Workstep for details. To change the Preferences for this application: 1. Select the Preferences Data Manager. 2. Click Open Preference Editor at the bottom of the Preferences section (the second icon: ). NM Application Suite Release 1.0 Cardiac 113

123 4.4 Review Layouts 3. Make changes in the preferences window using the information in the table below. See the section on Creating and Editing Preferences on page 43 for details on editing Preferences. You can save these parameters as Preferences: Parameter Default Description Review Compress Factor 1 Number of frames to compress for review data 4.4 Review Layouts Below are the layouts in the Review workstep: Muga Dynamic Muga Cine Shunt Dynamic Shunt Splash First Pass Splash First Pass Dynamic Gated FirstPass SC images Philips Healthcare 114 Cardiac NM Application Suite Release 1.0

124 5 Whole Body This application allows you to quantify multiple types of skeletal image data and review the results. It has these methods: Three Phase Analysis: This calculates count density mean ratios. Ileosacrum Ratio Computation: This calculates the ratio of the Left and Right Ileosacrum with respect to the Spine (Spine = 100%). For information on loading requirements, and on calculations and algorithms used in this application, see the appropriate section in the NM Application Suite Reference Manual. 5.1 Using Three Phase Analysis As you draw the Numerator/Denominator ROI pairs, the ratio is displayed in the viewer. Philips Healthcare Rev A Figure 48 Numerator/denominator ROI pair and ratio 5.2 Results for Three Phase Analysis Composite image with ROIs and Ratio Pool & Delayed images, with ROIs and Ratio NM Application Suite Release 1.0 Whole Body 115

125 5.3 Parameters for Three Phase Analysis Splash display with ROIs Time Activity curve for Flow Numerator and Denominator If you do not see all the result images in the Review Results workstep, it may be that one or more viewers are hidden. If you suspect this, try using the Show Hidden Viewers tool in the Utilities Data Manager. See the section on Review Results Workstep for details. 5.3 Parameters for Three Phase Analysis To change the Preferences for this application: 1. Select the Preferences Data Manager. 2. Click Open Preference Editor at the bottom of the Preferences section (the second icon: ). 3. Make changes in the preferences window using the information in the table below. See the section on Creating and Editing Preferences on page 43 for details on editing Preferences. You can save these parameters as Preferences: Parameter Default Description Smooth Curve False This determines whether the results curves are smoothed. 5.4 Using Ileosacrum Ratio Computation If you do not have the required dataset, you can create a substitute by displaying the Posterior image, adjusting the red limit bars, and using Extract in the right-click context menu. This allows you to proceed to the Define Regions workstep and draw ROIs. 5.5 Results for Ileosacrum Ratio Computation Review image with all ROIs Counts in Left, Right, and Spine regions Number of pixels in Left, Right, and Spine regions Normalized counts in Left, Right, and Spine regions Raw (%) for Left and Right compared to Spine Philips Healthcare 116 Whole Body NM Application Suite Release 1.0

126 Review Layouts 5.6 Normalized (%) for Left and Right compared to Spine L/R Raw Ratio L/R Normalized Ratio IS-Ratio L/R IS-Ratio R/L Left SIJ Index Right SIJ Index All Images If you do not see all the result images in the Review Results workstep, it may be that one or more viewers are hidden. If you suspect this, try using the Show Hidden Viewers tool in the Utilities Data Manager. See the section on Review Results Workstep for details. 5.6 Review Layouts Below are the layouts in the Review workstep: Whole Body Comparison Flow With Statics Static Review Whole Body With Spots Whole Body Display Fusion Display SC images Philips Healthcare Rev A NM Application Suite Release 1.0 Whole Body 117

127 5.6 Review Layouts Philips Healthcare 118 Whole Body NM Application Suite Release 1.0

128 6 Lung This application allows you to generate relative pulmonary uptake values for various pulmonary studies. Depending on the datasets you have loaded, you can perform analysis for perfusion only, ventilation only, or washout. Results are provided for upper, middle, and lower vertical segments in each lung. It has these methods: Perfusion Analysis Ventilation Analysis Washout Analysis For information on loading requirements, and on calculations and algorithms used in this application, see the appropriate section in the NM Application Suite Reference Manual. 6.1 Using Lung This application draws a bounding box around each lung. To use automatic edge detection, first adjust each bounding box by dragging its control points. Then click Detect All Regions. NOTE: If you have loaded both posterior and anterior images, you can redraw ROIs separately on each image. However, you must be sure to select the image first by clicking on it. 6.2 Washout Results In the Washout Analysis workstep, as soon as you draw the ROIs, you see the Results information: Philips Healthcare Rev A Cine with ROIs Static images T1/2 values for each segment in each lung Splash display with ROIs Time Activity curves for each segment in each lung If you do not see all the result images in the Review Results workstep, it may be that one or more viewers are hidden. If you suspect this, try using the Show Hidden Viewers tool in the Utilities Data Manager. See the section on Review Results Workstep for details. NM Application Suite Release 1.0 Lung 119

129 6.3 Ventilation Results 6.3 Ventilation Results Posterior and Anterior images with ROIs 6 viewers for static images in other orientations Posterior: For each lung, counts by segment, in counts and as a percentage Anterior: For each lung, counts by segment, in counts and as a percentage Geometric Mean: For each lung, counts by segment, in counts and as a percentage If you do not see all the result images in the Review Results workstep, it may be that one or more viewers are hidden. If you suspect this, try using the Show Hidden Viewers tool in the Utilities Data Manager. See the section on Review Results Workstep for details. 6.4 Perfusion Results Posterior and Anterior images with ROIs 6 viewers for static images in other orientations Posterior: For each lung, counts by segment, in counts and as a percentage Anterior: For each lung, counts by segment, in counts and as a percentage Geometric Mean: For each lung, counts by segment, in counts and as a percentage If you do not see all the result images in the Review Results workstep, it may be that one or more viewers are hidden. If you suspect this, try using the Show Hidden Viewers tool in the Utilities Data Manager. See the section on Review Results Workstep for details. 6.5 Review Layouts Below are the layouts in the Review workstep: Perfusion Ventilation Compare Perfusion Display Ventilation Display Washout Display Philips Healthcare 120 Lung NM Application Suite Release 1.0

130 Review Layouts 6.5 Washout Perfusion Display SC images Philips Healthcare Rev A NM Application Suite Release 1.0 Lung 121

131 6.5 Review Layouts Philips Healthcare 122 Lung NM Application Suite Release 1.0

132 7 Renal The Renal application allows you to analyze and display dynamic singleand multi-phase planar renal studies. IMPORTANT: The application uses the Pre Injection Syringe and Post Injection Syringe images, if loaded, to compute the injected dose. Use the following guidelines when acquiring pre- and post-injection syringe images: Philips Healthcare Rev A Place the syringe on the table 30 cm from the collimator face. Under the syringe, include all other items typically used between a patient and the table (e.g., sheet, pad, restraining strap). If the radiopharmaceutical activity of the pre-injection syringe is greater than 3 mci, use a syringe shield when acquiring images for both the pre- and the post-injection syringes. Pre-injection syringe activity less than 3 mci does not require shielding. If you typically image patients in the supine position, position the detector under the table. If you use a different patient position, position the detector accordingly. Acquire a one-minute static image of each syringe, using a 256 x 256 matrix to avoid pixel overflow. This application allows you to process and analyze planar renal studies of 1, 2, or 3 phases. You can use this application to: Load and process dynamic images from a renal study for display and processing. You can limit the set of frames in the loaded dynamic sets to a consecutive subset of frames for use in processing. Create your own predefined default sets and define the rules for matching the default sets to study types and protocols (with the appropriate system permissions). By default, the application uses predefined parameter values based on the protocol name of the selected study. Choose a processing algorithm. Choices include: ERPF (plasma and camera-based); GFR (plasma and camera-based); Lasix; Transplant; Gates; Taylor; Oberhausen; Bubeck; Tauxe; and UAB. The application takes the algorithm you choose as a default chosen based on exam name, and uses the algorithm to determine which ROIs are required. Inspect, define, and modify injection times and doses. Create composite images from the dynamic and use them for defining ROIs. Define pre- and post-injection syringe images and pre- and post-void bladder images. Perform automated organ segmentation, including identification of right and left kidneys, right and left background regions, and aorta as NM Application Suite Release 1.0 Renal 123

133 7.1 Using ROIs in Renal required by the processing algorithm chosen. You can choose the shape of the background ROIs. Manually define the required regions and modify any system-generated regions. After reviewing the curve plots of the specified regions, you can adjust the scale and displayed range of the plots. Perform Patlak corrections. Adjust the time interval for update calculations. Display a differential renogram. Compute and display the Hilson index for a transplant. View post-miction displays as applicable for the selected study. The application has these methods: Simple Renogram Pre Post Lasix Post Renogram Lasix GFR Gates ERPF Schlegel ERPF Schlegel with Void ERPF: MAG 3 Renal Deconvolution DMSA Static Ratio Hilson Index Patlak Cortical Analysis Split T1/2 For information on loading requirements, and on calculations and algorithms used in this application, see the appropriate section in the NM Application Suite Reference Manual. 7.1 Using ROIs in Renal The Renal application has some unique ROI functionality. While other applications may allow you to draw ROIs semi-automatically, manually, or by using a template, Renal also provides completely automatic ROI definition in some instances. To use this, select the pink area icon in the ROI pull-down menu for the ROI you need to set. Philips Healthcare 124 Renal NM Application Suite Release 1.0

134 Supplying Inputs for Renal 7.2 In the Define Regions workstep of some Preferences, this application draws a preliminary bounding box around one or more regions. You must then take these steps to create ROIs: 1. Move and adjust the bounding boxes so they just enclose the appropriate areas. Drag a line to move a bounding box; drag a handle to reshape it. 2. Click the Detect All Regions icon to create isocontours for all the regions at once. This creates ROIs for the areas and for any required backgrounds. To start over, click the Detect All Regions icon again; to draw the ROI by hand, click the eraser icon (Draw Region) in the ROI control. It turns into a pencil, indicating that you can draw manually. 7.2 Supplying Inputs for Renal Some Renal Preferences require inputs in the Define Regions workstep that affect the results calculations. There are two inputs that are only displayed in the results, and are not used for calculations. These are Radionuclide and Radiopharmaceutical. These are read from the DICOM header if they are present; if not, you can type any value for them. 7.3 Renal Results If you do not see all the result images in the Review Results workstep, it may be that one or more viewers are hidden. If you suspect this, try using the Show Hidden Viewers tool in the Utilities Data Manager. See the section on Review Results Workstep for details. Philips Healthcare Rev A 7.4 Preferences To change the Preferences for this application: 1. Select the Preferences Data Manager. 2. Click Open Preference Editor at the bottom of the Preferences section (the second icon: ). 3. Make changes in the preferences window using the information in the table below. See the section on Creating and Editing Preferences on page 43 for details on editing Preferences. You can save these parameters for this application: NM Application Suite Release 1.0 Renal 125

135 7.5 Review Layouts Parameter Default Description Review Compress Factor (Renogram) 1 Number of frames to compress for renogram review data Review Compress Factor (Flow) 1 Number of frames to compress for flow review data Composite Start Frame 1 First frame of composite image Composite End Frame (varies depending on the method) Last frame of composite image For more on compression, see the section on Frame Compress on page Review Layouts Below are the layouts in the Review workstep: Renal Dynamic Renal Review Static Review DMSA Static Review SC images 7.6 Simple Renogram Using Simple Renogram Simple Renogram is primarily used to analyze the acquisition data and produce statistical information about the function of the kidneys. In the Define Regions workstep, you may need to provide values for the following parameters in the Inputs viewer: Radionuclide Radiopharmaceutical Philips Healthcare 126 Renal NM Application Suite Release 1.0

136 Pre Post Lasix 7.7 Time of Lasix Injection (min) Differential Start Time (min) Differential End Time (min) Residual Activity Time (min) Curves Smoothing Factor Results Composite image with all ROIs Cine with all ROIs Peak Time (sec) for both kidneys T 1/2 (min) for both kidneys Peak Counts for both kidneys Diff Perfusion (%) for both kidneys Renal Retention (%) for both kidneys Differential Calculation Time (min) Radionuclide Radiopharmaceutical Time of Lasix Injection (min) Splash display Time Activity curve for Flow data for both kidneys and aorta Time Activity curve for Renogram data for both kidneys and bladder 7.7 Pre Post Lasix Philips Healthcare Rev A Using Pre Post Lasix This method provides renal processing capability (analogous to Pegasys) by combining the flow and renal portions of the study into a single curve for processing. In the Define Regions workstep, you may need to provide values for the following parameters in the Inputs viewer: Radionuclide Radiopharmaceutical NM Application Suite Release 1.0 Renal 127

137 7.7 Pre Post Lasix Time of Lasix Injection (0 for no Lasix) Time of Post Lasix Percent Remaining (min) Differential Start Time (min) Differential End Time (min) Residual Activity Time (min) Curves Smoothing Factor NOTE: If Time of Lasix Injection is blank or zero, it will not appear in the Review Results workstep Results Composite image with all ROIs Cine with all ROIs Time Activity curves for Flow data for both kidneys and aorta Time Activity curves for Renogram data for both kidneys and bladder Splash display Renogram Results Peak Time (min) T 1/2 (min) Peak Counts 20 Min / Peak Ratio (%) Differential (%) Flow Image Timings Lasix Clearance Report Time To Max (sec) Max Flow Counts Peak Post Lasix (min) T 1/2 From Peak (min) T 1/2 from Lasix (min) PRE-L/10M Clearance (%) PK-L/10M Clearance (%) Aorta Peak Time (sec) Philips Healthcare 128 Renal NM Application Suite Release 1.0

138 Post Renogram Lasix 7.8 Differential Time (min) Radionuclide Radiopharmaceutical Time of Lasix Injection (min) 7.8 Post Renogram Lasix This method uses a post renogram Single Framing Rate Dynamic (Diuresis) study to create the Post-Lasix T 1/ Using Post Renogram Lasix In the Define Regions workstep, you may need to provide values for the following parameters in the Inputs viewer: Radionuclide Radiopharmaceutical Curves Smoothing Factor Time of Lasix Injection (min) Results Philips Healthcare Rev A Composite image with all ROIs Cine with all ROIs T 1/2 (min) for both kidneys Radionuclide Radiopharmaceutical Time of Lasix Injection (min) Time Activity curve for left kidney Lasix and fit Time Activity curve for right kidney Lasix and fit Splash display NM Application Suite Release 1.0 Renal 129

139 7.9 GFR Gates 7.9 GFR Gates This calculates the total glomerular filtration rate (GFR), individual GFR, normalized GFR, time to peak, and time to half peak for two-kidney, single-kidney, and kidney transplant patients Using GFR Gates In the Define Regions workstep, you may need to provide values for the following parameters in the Inputs viewer: Radionuclide Radiopharmaceutical Patient Weight (lb) Patient Height (in) Pre Dose Multi-Factor Post Dose Multi-Factor Pre-Injection Counts Post-Injection Counts Pre-Injection Time Patient Injection Time Post-Injection Time Total Injected Dose Results Composite image with all ROIs Cine with all ROIs Depth (cm) for both kidneys Peak Time (sec) for both kidneys T 1/2 (min) for both kidneys Uptake (%) for both kidneys GFR (ml/min) for both kidneys Total GFR (ml/min) Normalized GFR (ml/min) Total Injected Dose (MBq) Philips Healthcare 130 Renal NM Application Suite Release 1.0

140 ERPF Schlegel 7.10 Radionuclide Radiopharmaceutical Splash display Time Activity curves for Flow data for both kidneys and aorta Time Activity curves for Renogram for both kidneys and bladder 7.10 ERPF Schlegel This method calculates the effective renal plasma flow Using ERPF Schlegel In the Define Regions workstep, you may need to provide values for the following parameters in the Inputs viewer: Radionuclide Radiopharmaceutical Patient Weight (lbs) Patient Height (in) Total Injected Dose (MBq) Pre-Injection Counts Post-Injection Counts Results Philips Healthcare Rev A Schlegel Results Composite image including all ROIs Cine including all ROIs Time Activity curve for right and left kidney values, and bladder values Splash display Peak Time (sec) for both kidneys T 1/2 (min) for both kidneys Slope1 (peak - t1/2) for both kidneys Uptake (%) for both kidneys ERPF (ml/min) for both kidneys NM Application Suite Release 1.0 Renal 131

141 7.11 ERPF Schlegel with Void Normal Values Normalized ERPF (ml/min) for both kidneys, and total Total Values Total ERPF (ml/min) Normalized ERPF (ml/min) Radionuclide Radiopharmaceutical 7.11 ERPF Schlegel with Void This method calculates the effective renal plasma flow with additional calculation of dose return and residual urine volume Using ERPF Schlegel with Void In the Define Regions workstep, you may need to provide values for the following parameters in the Inputs viewer: Radionuclide Radiopharmaceutical Patient Weight (lbs) Patient Height (in) Voided Urine Volume (ml) Voided Urine Activity (MBq) Total Injected Dose (MBq) Pre-Injection Counts Post-Injection Counts Results Composite image including all ROIs Cine including all ROIs Time Activity curve for right and left kidney values, and bladder values Splash display Philips Healthcare 132 Renal NM Application Suite Release 1.0

142 ERPF MAG Schlegel with Void Results Peak Time (sec) for both kidneys T 1/2 (min) for both kidneys Slope1 (peak - t1/2) for both kidneys Uptake (%) for both kidneys ERPF (ml/min) for both kidneys Normal Values Normalized ERPF (ml/min) for right and left kidneys Total Values Total ERPF (ml/min) Normalized ERPF (ml/min) Return Values Residual Return (%) Voided Return (%) Predicted Return (%) Normal Return (%) Total Return (%) Radionuclide Radiopharmaceutical 7.12 ERPF MAG 3 Philips Healthcare Rev A Using ERPF MAG 3 The method implements calculation of effective renal plasma flow from MAG-3. In the Define Regions workstep, you may need to provide values for the following parameters in the Inputs viewer: Radionuclide Radiopharmaceutical Patient Weight (lbs) Patient Height (in) NM Application Suite Release 1.0 Renal 133

143 7.13 Deconvolution Pre-Injection Counts Post-Injection Counts Pre Dose Multi-Factor Post Dose Multi-Factor Results Composite image with all ROIs Cine with all ROIs T 1/2 (min) for both kidneys Peak Time (min) for both kidneys Uptake (%) for both kidneys ERPF (ml/min) for both kidneys Total MAG3 Clearance (ml/min) Total ERPF (ml/min) Radionuclide Radiopharmaceutical Time Activity Flow curves for both kidneys Time Activity Renogram curves for both kidneys Splash display 7.13 Deconvolution This method allows you to generate kidney retention functions with transit calculation Using Renal Deconvolution In the Define Regions workstep, you may need to provide values for the following parameters in the Inputs viewer: Radionuclide Radiopharmaceutical Time of Lasix Injection (min) Philips Healthcare 134 Renal NM Application Suite Release 1.0

144 DMSA Static Ratio Results Composite image with all ROIs Cine with all ROIs Mean Transit Time (min) for both kidneys T80 (min) for both kidneys T20 (min) for both kidneys % Function (%) Peak Time (min) for both kidneys Radionuclide Radiopharmaceutical Time of Lasix Injection (min) Splash display Time Activity background subtracted curves for Renogram data for both kidneys and blood Time Activity curves for Function data for left and right retention function 7.14 DMSA Static Ratio This method provides 2 ROI ratio comparison (optionally with background subtraction) analogous to the Pegasys Global Q implementation Results Philips Healthcare Rev A Posterior image with ROIs Counts/Total Vol for both kidneys %Diff/Total Vol for both kidneys Counts/Unit Vol for both kidneys %Diff/Unit Vol for both kidneys Geometric Mean Counts for both kidneys Any optional images loaded If you do not see all the result images in the Review Results workstep, it may be that one or more viewers are hidden. If you suspect this, try using the Show Hidden Viewers tool in the Utilities Data Manager. See the section on Review Results Workstep for details. NM Application Suite Release 1.0 Renal 135

145 7.15 Hilson Index 7.15 Hilson Index The Hilson Index method generates a renal perfusion index based on the ratio of areas under the kidney curve and the area under the artery curve in the perfusion phase. It is typically used for kidney transplant management Using Hilson Index After you have drawn the ROIs, you can optionally enter these values in the Inputs viewer: Radionuclide Radiopharmaceutical Results Composite image with all ROIs Cine with all ROIs Peak Time (min) T 1/2 (min) Aorta Peak Time (sec) Area Aorta Curve To Peak (cps) Area Renal Curve (cps) Hilson Perfusion Index Peak Counts (cps) Up Slope (cpm) Rise Time (min) 20 min / Peak Ratio (%) 20 min/ 3 min Ratio (%) EOS (cps) TEOS (min) Total Counts (cps) Total Counts Range (2 to 3 mins) (cps) Radionuclide Radiopharmaceutical Time Activity curve for Perfusion data for aorta and transplant kidney Philips Healthcare 136 Renal NM Application Suite Release 1.0

146 Patlak 7.16 Time Activity curve for Clearance data for aorta and transplant kidney Splash display 7.16 Patlak This method provides the capability to generate Patlak-Rutland plots and display separate tissue and vascular background regions together with a Patlak-Rutland plot (Analogous to the Odyssey Euro Custom Menu) Using Patlak In the Define Regions workstep, you may need to provide values for the following parameters in the Inputs viewer: Radionuclide Patient Weight (lbs) Time of Lasix Injection (min) Activity Administered (MBq) Results Philips Healthcare Rev A Composite image with all ROIs Cine with all ROIs Peak Time (sec) for both kidneys Relative Function (%) for both kidneys RF Time Range (sec) for both kidneys Two Minute Uptake (%) for both kidneys Three Minute Uptake (%) for both kidneys 20 Min/Peak Ratio (%) for both kidneys Radionuclide Radiopharmaceutical Time of Lasix Injection (min) Splash display Rutland Fit curves for each kidney Time Activity curves for both kidneys and bladder You can use the timing markers in the Time Activity graph to adjust the fit curves, which also causes the results values to change. NM Application Suite Release 1.0 Renal 137

147 7.17 Cortical Analysis 7.17 Cortical Analysis This method computes and displays various results using the Cortical regions Using Cortical Analysis In the Define Regions workstep, you may need to provide values for the following parameters in the Inputs viewer: Radionuclide Radiopharmaceutical Time of Lasix Injection (min) RCA Calculation Time (min) Curves Smoothing Factor Results 7.18 Split T 1/2 Composite image with all ROIs Cine with all ROIs T 1/2 (min) for both cortexes Peak Time (min) for both cortexes Peak Counts for both cortexes Residual Cortical Activity (%) RCA Calculation Time (min) Radionuclide Radiopharmaceutical Time of Lasix Injection (min) Time Activity curve for Renogram data for both cortexes Splash display Philips Healthcare This method uses a post-renogram single framing rate dynamic to create separate right and left T 1/2 curves with associated control timing markers and appropriate curve fitting. 138 Renal NM Application Suite Release 1.0

148 Split T 1/ Using Split T1/2 In the Define Regions workstep, you may need to provide values for the following parameters in the Inputs viewer: Radionuclide Radiopharmaceutical Curves Smoothing Factor In the Review Results workstep, you can use the timing markers in the curve viewer to adjust the time range represented by the results values. Drag the timing markers to update the results Results Composite image with all ROIs Cine with all ROIs T 1/2 (min) for both kidneys Radionuclide Radiopharmaceutical Time Activity curve and markers, with fit for left kidney Time Activity curve and markers, with fit for right kidney Splash display Philips Healthcare Rev A NM Application Suite Release 1.0 Renal 139

149 7.18 Split T 1/2 Philips Healthcare 140 Renal NM Application Suite Release 1.0

150 8 Endocrine This application allows you to process and review thyroid and parathyroid imaging studies. Using this application, you can display and analyze both dual isotope (Tc-Tl) and delayed Mibi parathyroid studies, as well as thyroid studies, including calculation of differential uptakes. It has these methods: Parathyroid Subtraction: This subtracts the technetium image from the normalized thallium image and displays a subtracted image. This is the Dual Isotope technique. Parathyroid Dual-Phase: This method allows you to compare the phases in dual-phase data. It is for display only, and has no ROIs or results. Thyroid: Thyroid uptake is calculated in the total area in up to six regions of interest. Full and Empty Syringe data from a gamma-camera image or Dose Calibrator input are used. The size of the thyroid is calculated in cm 2, weight in grams, and volume in cm 3. For information on loading requirements, and on calculations and algorithms used in this application, see the appropriate section in the NM Application Suite Reference Manual. 8.1 Thyroid Thyroid uptake is calculated in the total area in up to six regions of interest. Full and Empty Syringe data from gamma-camera image or Dose Calibrator input are taken. Size of the Thyroid is calculated in cm 2, weight in grams and Volume in cm Using Thyroid Philips Healthcare Rev A In the Define Regions workstep, you may need to provide values for the following parameters in the Inputs viewer: Select Radionuclide Additionally, if the Full or Empty syringe images are not present, you may need to provide the following parameters, which are used in the calculation of thyroid's percent uptake: Assayed Dose - Pre Administration (MBq or MCi) Assayed Dose - Post Administration (MBq or MCi) Calibration Factor (cpm/kbq or MCi) NM Application Suite Release 1.0 Endocrine 141

151 8.1 Thyroid Pre Administration Measurement Time (M/d/yyyy h:mm:ss tt) Post Administration Measurement Time (M/d/yyyy h:mm:ss tt) Thyroid Measurement Time (M/d/yyyy h:mm:ss tt) Calibration Factor: If Syringe or capsule data is used from a Dose Calibrator, a calibration factor must be determined to establish the relation between counts per minute and kilobequerels (or microcuries). The calibration factor is determined using a known amount of activity in MBq (e.g. 37MBq or 1 mci). NOTE: By default, the EBW system is set to use Bq. To change to Ci, open the EBW Preferences and change it in the PET Preferences Viewing page. If you use mci, convert uci for kbq in this example. To calculate the Calibration Factor: 1. Assay the capsule (or syringe) in a dose calibrator and record the result. 2. Acquire a static image of the capsule in a thyroid phantom using the same collimator and distance from the collimator as that used for a patient. 3. Load the capsule image in the General Review application by selecting it and clicking NM Application Suite in the EBW Review panel. 4. Select a tool from the Measure list (right-click on the image or go to the Utilities manager). 5. Draw a tight ROI around the capsule. 6. Right-click on the ROI and select Total Value from the menu. 7. Note the number displayed for the ROI and convert it to counts per minute. 8. Divide the counts per minute by activity as expressed in kilobequerels or microcuries (e.g kbq or 1000 uci) to obtain the calibration factor in cpm/kbq or cpm/uci. NOTE: The calibration factor must be determined again if different collimators are used or when the gamma camera has been tuned Results Thyroid image with ROIs Counts (in the thyroid region and any optional regions) Number of Pixels (in the thyroid region and any optional regions) Uptake (%) (in the thyroid region and any optional regions) Philips Healthcare 142 Endocrine NM Application Suite Release 1.0

152 Parathyroid 8.2 If you do not see all the result images in the Review Results workstep, it may be that one or more viewers are hidden. If you suspect this, try using the Show Hidden Viewers tool in the Utilities Data Manager. See the section on Review Results Workstep for details. 8.2 Parathyroid The Subtraction method subtracts the thyroid image from the normalized parathyroid image and displays a subtracted image. This is the Dual Isotope technique. You can use Mibi and delayed Mibi or Tl and Tc images Using Parathyroid In the Define Regions workstep, you are required to draw a background region first, and then the thyroid region. Usually the order is reversed, but for this application, drawing the background first allows the software to create a background-subtracted image so that the organ is better defined. This means that you can draw a more accurate ROI for the thyroid. If the two images are not aligned and motion correction is required, draw the ROIs as usual on the Thyroid image. Then drag and adjust the ROI on the Parathyroid image so the ROI is correctly positioned around the organ. This will align the images before subtraction Subtraction Results Philips Healthcare Rev A 8.3 Review Layouts The Tl and Tc images side-by-side with the defined region overlaid on both images; each image also displays the total count at the bottom The subtracted image; use the bar at the bottom to set the Subtraction Factor (range = 0-3) 8 subtracted images with subtraction factors of 3.0F, 2.0F, 1.5F, 1.0F, 0.9F, 0.8F, 0.7F, 0.6F If you do not see all the result images in the Review Results workstep, it may be that one or more viewers are hidden. If you suspect this, try using the Show Hidden Viewers tool in the Utilities Data Manager. See the section on Review Results Workstep for details. Below are the layouts in the Review workstep: Parathyroid Side by Side NM Application Suite Release 1.0 Endocrine 143

153 8.3 Review Layouts Parathyroid Static Review Thyroid Static Review SC images Philips Healthcare 144 Endocrine NM Application Suite Release 1.0

154 9 Esophagus This application generates a set of time-activity curves for a dynamic esophageal study, based on an ROI that you define. Typically, it is done on patients who have difficulty in swallowing, have gastro-esophageal reflux, or have peristalsis problems. It has these methods: Esophagus Gastro-Esophagus Reflux For information on loading requirements, and on calculations and algorithms used in this application, see the appropriate section in the NM Application Suite Reference Manual. 9.1 Using Esophagus By default, this application draws a rectangular bounding box with three sections around the esophagus. You can adjust the box using its control points, or change the drawing mode to manual to use a different drawing shape. 9.2 Esophagus This allows you to get esophageal transit information by segment, including time activity curves, empty ratio, and peak ratio Using Esophagus Philips Healthcare Rev A Esophagus Results In the Review Results workstep, you can use the timing markers in the curve viewer to adjust the time range represented by the results values. Drag the timing markers to update the results. Cine with all ROIs Time Activity curve for each of the ROI segments Empty Ratio for each segment and the whole esophagus Peak Ratio for each segment and the whole esophagus Splash display of the images Time Activity curve for the whole NM Application Suite Release 1.0 Esophagus 145

155 9.3 Gastro-Esophagus Reflux A condensed image for each dynamic dataset, in which each column is the sum across the y-axis of each frame of the dataset; the columns are presented side-by-side in frame order. If you do not see all the result images in the Review Results workstep, it may be that one or more viewers are hidden. If you suspect this, try using the Show Hidden Viewers tool in the Utilities Data Manager. See the section on Review Results Workstep for details Preferences To change the Preferences for this application: 1. Select the Preferences Data Manager. 2. Click Open Preference Editor at the bottom of the Preferences section (the second icon: ). 3. Make changes in the preferences window using the information in the table below. See the section on Creating and Editing Preferences on page 43 for details on editing Preferences. You can save these parameters for this application: Parameter Default Description Review Compress Factor 1 Compression value for review data 9.3 Gastro-Esophagus Reflux This allows you to get reflux information, including empty and peak ratios, and a time activity curve Using Gastro-Esophagus Reflux Gastro-Esophagus Reflux Results In the Review Results workstep, you can use the timing markers in the curve viewer to adjust the time range represented by the results values. Drag the timing markers to update the results. Cine with all ROIs Time Activity curve for the whole Splash display of the images Philips Healthcare 146 Esophagus NM Application Suite Release 1.0

156 Review Layouts 9.4 Empty Ratio Peak Ratio A condensed image for each dynamic dataset, in which each column is the sum across the y-axis of each frame of the dataset; the columns are presented side-by-side in frame order. If you do not see all the result images in the Review Results workstep, it may be that one or more viewers are hidden. If you suspect this, try using the Show Hidden Viewers tool in the Utilities Data Manager. See the section on Review Results Workstep for details Preferences To change the Preferences for this application: 1. Select the Preferences Data Manager. 2. Click Open Preference Editor at the bottom of the Preferences section (the second icon: ). 3. Make changes in the preferences window using the information in the table below. See the section on Creating and Editing Preferences on page 43 for details on editing Preferences. You can save these parameters for this application: Parameter Default Description Review Compress Factor 1 Number of frames to compress for review data 9.4 Review Layouts Philips Healthcare Rev A Below are the layouts in the Review workstep: Splash Display SC images NM Application Suite Release 1.0 Esophagus 147

157 9.4 Review Layouts Philips Healthcare 148 Esophagus NM Application Suite Release 1.0

158 10 Gastro Intestinal You can use a Linear, Elashoff, or Siegel curve fit method for these calculations. You can also compute a new series of frames equal to the geometric mean of corresponding anterior and posterior series (if you have loaded these series).this application provides data on the clearance time of stomach contents and calculates the T1/2 for gastric emptying. The application supports two isotope image sets: Tc for Solid and In111 for Liquid emptying calculations. A linear fit curve is used for Solid data and an exponential fit curve for Liquid. When both series are loaded, a new series of frames equal to the geometric mean is computed. It has these methods: Solid Liquid For information on loading requirements, and on calculations and algorithms used in this application, see the appropriate section in the NM Application Suite Reference Manual Using Gastric Emptying In the Setup workstep, you can load multiple dynamic or static datasets. The application will append datasets according to the time value in the DICOM header. In the Review Results workstep, you can use the timing markers in the curve viewer to adjust the time range represented by the results values. Drag the timing markers to update the results Results Philips Healthcare Rev A Depending on the Preference, the following results are available. Anterior composite image with ROIs Posterior composite image with ROIs Cine loop for Anterior with ROIs overlaid on all images Cine loop for Posterior with ROIs overlaid on all images Splash display for Anterior with ROIs Splash display for Posterior with ROIs Time to Half NM Application Suite Release 1.0 Gastro Intestinal 149

159 10.3 Preferences Time to Peak Retention at 30, 60, 90, 120, and 240 minute lag times End Time Retention Max and Min Line Retentions Time Activity curve for each ROI Time Activity viewer containing the Fit curve and decay corrected Residual curve NOTE: If you have loaded geometric mean data, the Residual curve is actually the geometric mean curve. If you do not see all the result images in the Review Results workstep, it may be that one or more viewers are hidden. If you suspect this, try using the Show Hidden Viewers tool in the Utilities Data Manager. See the section on Review Results Workstep for details Preferences To change the Preferences for this application: 1. Select the Preferences Data Manager. 2. Click Open Preference Editor at the bottom of the Preferences section (the second icon: ). 3. Make changes in the preferences window using the information in the table below. See the section on Creating and Editing Preferences on page 43 for details on editing Preferences. You can save these parameters for this application: Parameter Default Description Review Compress Factor 1 Number of frames to compress for review data Smooth Curve False This determines whether the Residual Curve is smoothed. Solid or Liquid, depending on the Preference selected - Decay Corrected Curve True This determines whether the Decay Corrected curve appears in the Time Activity viewer. Curve for Calculation Residual Curve This determines whether the Fit or Residual curve is used to calculate all the Results. Philips Healthcare 150 Gastro Intestinal NM Application Suite Release 1.0

160 Review Layouts Review Layouts Below are the layouts in the Review workstep: Gastro Intestinal Dynamic Review Gastro Intestinal Splash Gastro Intestinal Static Review SC Images Philips Healthcare Rev A NM Application Suite Release 1.0 Gastro Intestinal 151

161 10.4 Review Layouts Philips Healthcare 152 Gastro Intestinal NM Application Suite Release 1.0

162 11 Hepatobiliary This application allows you to perform gall bladder static analysis and gall bladder dynamic analysis. It has these methods: Gall Bladder: This displays a dynamic series of gallbladder images and calculates the gallbladder ejection fraction. It involves no further radionuclide injections, but requires a timed cholecystokinin octapeptide (CCK) infusion part way through the study. You can get results for these ROIs, which are also selectable as methods for Gall Bladder: Gall Bladder (required) BackGround Hepatic Duct Common Bile Duct Duodenum GBEF Static Analysis: This calculates the percent emptied from the pre-intervention (CCK) Gallbladder static and up to 12 Post-intervention statics. GBEF Static Analysis: This calculates the percent emptied from the pre-intervention (CCK) Gallbladder static and up to 12 Post-intervention statics. For information on loading requirements, and on calculations and algorithms used in this application, see the appropriate section in the NM Application Suite Reference Manual Using GallBladder Philips Healthcare Rev A In the Define Regions workstep, a splash display is present for reference, and a Time Activity curve is drawn as soon as all required ROIs are present. Additionally, you may need to provide values for the following parameters in the Inputs viewer: Stimulus Info Dosage of CCK Time CCK Infusion Began Duration of CCK infusion Morphine Administered NM Application Suite Release 1.0 Hepatobiliary 153

163 11.2 Results for GallBladder In the Review Results workstep, you can use the timing markers in the Gall Bladder curve viewer to adjust the time range represented by the results values. By default, the green line indicates maximum counts, and the red line indicates minimum counts. Drag the timing markers to move them Results for GallBladder The Gallbladder Ejection Fraction application displays a dynamic series of gallbladder images and calculates the gallbladder ejection fraction. Depending on the preference selected, the results include a combination of the following: Cine loop with all ROIs The Ejection Fraction GBEF (%), based true counts, and on the timing markers in the time activity curve The Ejection Period GBEP (min) based on the timing markers in the time activity curve The Ejection Rate GBER (%/min) (time between injection and begin of emptying) based on the timing markers in the time activity curve Maximum Counts based on the timing markers in the time activity curve Minimum Counts based on the timing markers in the time activity curve Dosage of CCK (or morphine, if selected) Time CCK (or morphine, if selected) Infusion Began Duration of CCK (or morphine, if selected) infusion Latent Period GBLP (min) Morphine Administered Time Activity curve for background corrected Gall Bladder Time Activity curve for background corrected Hepatic Duct Time Activity curve for background corrected Common Bile Duct Time Activity curve for background corrected Duodenum Splash Display If you do not see all the result images in the Review Results workstep, it may be that one or more viewers are hidden. If you suspect this, try using the Show Hidden Viewers tool in the Utilities Data Manager. See the section on Review Results Workstep for details. Philips Healthcare 154 Hepatobiliary NM Application Suite Release 1.0

164 Preferences for GallBladder Preferences for GallBladder To change the Preferences for this application: 1. Select the Preferences Data Manager. 2. Click Open Preference Editor at the bottom of the Preferences section (the second icon: ). 3. Make changes in the preferences window using the information in the table below. See the section on Creating and Editing Preferences on page 43 for details on editing Preferences. You can save these parameters for this Preference: Parameter Default Description Review Compress Factor 1 Number of frames to compress for review data Stimulus Info NoStimulus This is the type of stimulus used in the study. Decay Corrected Curve Yes This determines whether the Time Activity curve uses decay correction data Results for GBEF Static Analysis If multiple Post Stimulus images are loaded: Philips Healthcare Rev A Counts and % Emptied values for PreStimulus Counts and % Emptied values for all Post Stimulus data If one Post Stimulus image is loaded: PreStimulus counts PostStimulus counts Ejection Fraction value 11.5 Preferences for GBEF Static Analysis To change the Preferences for this application: 1. Select the Preferences Data Manager. 2. Click Open Preference Editor at the bottom of the Preferences section (the second icon: ). NM Application Suite Release 1.0 Hepatobiliary 155

165 11.6 Review Layouts 3. Make changes in the preferences window using the information in the table below. See the section on Creating and Editing Preferences on page 43 for details on editing Preferences. You can save these parameters for this Preference: Parameter Default Description Review Compress Factor 1 Number of frames to compress for review data 11.6 Review Layouts Below are the layouts in the Review workstep: GBEF Dynamic Display GBEF Splash Display SC images Static Review Philips Healthcare 156 Hepatobiliary NM Application Suite Release 1.0

166 12 General Review When you are in the EBW Patient Directory, if you click the NM Application Suite button in either the Review or Analysis panel, the selected studies open in the General Review application. This allows you to view the contents of a study without loading it into a processing application. It has no processing capabilities or results, and no preferences. However, all the functionality of the Control Panel on the left is still available, including the viewer tools and ICMT, so you can still perform operations on images. It also provides some display layouts, which vary according to the data that is loaded: All Image Display Whole Body Display Whole Body with Spots All Dynamic Display All Dynamic with Spots SPECT 3 View Orthogonal View Fusion Display Series Display Gated Planar Whole Body Intensity Compare Philips Healthcare Rev A NM Application Suite Release 1.0 General Review 157

167 158 General Review NM Application Suite Release 1.0 Philips Healthcare

168 13 QC Tools The QC Tools allow you to calculate the uniformity of an image. The result is two images, which you can then compare to check for uniformity. It uses static planar NM DICOM images as input. NOTE: Although the Application Palette appears as a Data Manager in the QC Tools application, it is empty. You cannot switch from the QC Tools to another application Using QC Tools Before using the QC Tools application, make sure you have an appropriate QC image, then follow the instructions below: 1. Click to highlight the patient dataset study you want to view. 2. Click on the Analysis icon s drop-down button. 3. Select the QC Tools icon from the Analysis applications. The QC Tools window appears. 4. Click on an image in the Pictorial Index (lower left). 5. The QC Tools application determines and displays the UFOV shape, however if you want to modify this, use the UFOV Type drop-down. Philips Healthcare Rev A NM Application Suite Release 1.0 QC Tools 159

169 13.2 UFOV and CFOV Results Tabs 6. Click OK. The loaded image, a smoothed version (bottom image), and its UFOV and CFOV results appear: Figure 49 QC Tools main window 13.2 UFOV and CFOV Results Tabs After an image loads, the center section of the main QC Tools window displays the following UFOV and CFOV image results: Minimum (Counts and Location) Maximum (Counts and Location) Row Differential Uniformity (Percentage) Column Differential Uniformity (Percentage) Max Row (Difference and Location) Max Column (Difference and Location) Philips Healthcare 160 QC Tools NM Application Suite Release 1.0

170 Using the QC Tools Toolbar Using the QC Tools Toolbar Each UFOV and CFOV viewer features a toolbar to customize your image. To use the tools in the toolbar, click on the toolbar icon (#1 below). A toolbar appears on the left with the controls described below. Figure 50 Viewer window with toolbar (left column) 1. Toolbar 2. Select Color Map: Use the drop-down menu to select a color map. 3. Invert Gray Level: Click to invert the image s pixel values. 4. Contrast Stretch: This sets the minimum and maximum pixel values to 0 and 255, adjusting the other values accordingly. 5. Text Box: Use this icon to add text to your image. Click where you want the text box to appear, and then type in your text. Click outside the text box to save it Saving Results Using the controls below, you can save results to text files, and you can save the display as a secondary capture or save it to film. Philips Healthcare Rev A Figure Saving to a Text File Save (left), Save current display (middle) and Film Display (right) Use Save to save the results to a text file. This displays the Enter Details dialog. 1. Some fields in the Results section are filled in by default; edit them if necessary: NM Application Suite Release 1.0 QC Tools 161

171 13.4 Saving Results Collimator Camera Test Conductor Date Pixel Size Detectors 2. Use the Store Results section to specify locations for the Summary Results and Additional Results. 3. To save the additional results, check Save Additional Results. 4. Type any other information into the Comments section. 5. Click Save Results Saving as Secondary Capture Use Save current display to save a multi-frame secondary capture of your QC Tools results. This displays the Save Secondary Capture Dialog. 1. Use the Save As pull-down menu to select an image format. Single-Frame Secondary Capture JPEG Multi-Frame Secondary Capture AVI 2. Type a Description and choose the appropriate settings. 3. For multi-frame and AVI formats, you can check Frame and specify the Start and End frames, and the number of frames to Skip. 4. For gated formats, you can check Bin and specify the Start and End bins, and the number of bins to Skip. 5. Check RGB to save color images, or Grayscale for grayscale images. 6. Click Save to save the images to the EBW Patient Directory. (Be sure to refresh the browser to see the new images.) Saving to Film Use Film Display (see Figure 51 above) to capture multi-image (full study window) captures. This creates an image that includes both the results viewer and all image viewers. NOTE: You can view and print the image using the EBW Film feature. Refer to your EBW Instructions for Use for details. Philips Healthcare 162 QC Tools NM Application Suite Release 1.0

172 14 Astonish Reconstruction If you have the Astonish license, Astonish appears as one of the reconstruction methods available. The Astonish method consists of performing Ordered Subsets Expectation Maximization with compensation for the blurring effects of your collimator built into the reconstruction. We frequently refer to this compensation as Resolution Recovery, because it allows the recovery of some of the original resolution of the activity distribution. To model the point spread of the activity distribution at the time of acquisition, Astonish uses the distance from the detector to the object of interest recorded as a function of angle by your camera during acquisition, and the geometric properties of the specific collimator used When to Use Astonish The accurate modeling performed by Astonish leads to excellent image quality for all images. Philips recommends using this method for most nuclear medicine SPECT reconstruction. Using the Astonish SPECT reconstruction method will enhance the resolution of your SPECT images, and improve the signal-to-noise ratio. The improved noise properties and appearance of the background of your images may change the appearance of the images. You may need to read several Astonish images in order to become comfortable with the appearance of these reconstructed images. The image below shows a patient study reconstructed according to typical clinical practice (bottom), and the same study reconstructed with Astonish (top). The improved resolution results in clearer separation and detail in the vertebrae. This is especially apparent in fine structures such as the breastbone. Philips Healthcare Rev A NM Application Suite Release 1.0 Astonish Reconstruction 163

173 14.2 Using Astonish Reconstruction Figure 52 A comparison of Astonish reconstruction (top) with FBP reconstruction (bottom) IMPORTANT: When using Astonish reconstruction for Cardiac images, use an Astonish database for your quantitation. AutoQUANT 7.0 provides an Astonish database, or you can use its database tools to generate normals databases specific to your site, your imaging preferences, and Astonish reconstruction with your preferred parameters (filtering, AC, scatter correction, etc.) Using Astonish Reconstruction Filtering and Noise You use Astonish in the same way as the other reconstruction methods: by selecting it from the Reconstruction pull-down menu or by using Astonish defaults. This section describes how to set reconstruction parameters such as Filtering, Iterations, and Subsets in order to optimize image quality when you use Astonish. With many iterative reconstruction methods, the noise in the reconstructed image increases with iterations. For this reason, the number of iterations is kept small to avoid having an overly noisy image. Astonish incorporates collimator effects so as to prevent this accumulation of noise, allowing you to use more total iterations while maintaining an acceptable image. Philips Healthcare 164 Astonish Reconstruction NM Application Suite Release 1.0

174 Using Astonish Reconstruction 14.2 However, if your images are noisy to begin with for instance, for gated cardiac datasets, rest perfusion imaging, or for general nuclear images acquired a long time after injection it is helpful to pre-filter the data both prior to and during reconstruction. For this reason, Philips provides you with a Hanning pre-filter to use with Astonish for noisy data. The two images below show the effect of filtering on an Astonish reconstructed image. The left image shows a Maximum Intensity Projection (MIP) display of an Astonish image reconstructed with 4 iterations and 16 subsets, with no pre-filter. The right image shows a similar display of an image reconstructed with the same parameters, but after applying a Hanning pre-filter with a cutoff of 1.0. The image has become smoother without losing the fine detail available as a result of the Astonish reconstruction. Figure 53 Astonish reconstruction without Hanning filter (left) and with Hanning filter (right) Iterations and Subsets Philips Healthcare Rev A To take full advantage of the resolution recovery provided with Astonish SPECT reconstruction, the expectation maximization algorithm must converge, which requires a large number of updates to the activity estimation for noisy data. Whereas for MLEM the number of updates is identical to the number of iterations, for OSEM, the subsets and iterations are multiplied together to calculate the total number of updates. The higher this number of updates is, the more likely the software is to have achieved convergence. Although it is possible to achieve an acceptable nuclear medicine reconstructed image using 2-3 iterations and 8-16 subsets, even better image quality may be achieved by iterating more. This may cause processing times to be extended, so your site must determine the acceptable number of iterations. While there is no maximum number of iterations, the maximum number of subsets is the number of projections in the data. Philips recommends starting from a Uniform estimate and using at least 24 updates for most nuclear medicine data, which can be achieved by performing 3 iterations with 8 subsets. NM Application Suite Release 1.0 Astonish Reconstruction 165

175 14.3 Further Reading The two images below show the effect of iteration number on an Astonish reconstructed image. The left image shows a Maximum Intensity Projection (MIP) display of an Astonish image reconstructed with 2 iterations and 16 subsets, with no pre-filter. The center image was reconstructed with 3 iterations, and the far right image with 4 iterations. More fine detail becomes available with additional iterations, but the images also become noisier. Figure 54 Astonish reconstruction with 2 iterations (left), 3 iterations (center), and 4 iterations (right) Applying Other Corrections As with other iterative methods, Astonish allows you to apply attenuation and scatter correction during reconstruction when transmission data is used along with the usual emission data. AutoSPECT Pro accepts CT data, Vantage transmission data, or a previously generated attenuation map to use in performing these corrections Further Reading Information on OSEM reconstruction To learn more about reconstruction using Resolution Recovery and other corrections, you may wish to read the following abstracts and papers: Almquist H, Arheden H, Arvidsson AH, Pahlm O, and Palmer J. Clinical implication of down-scatter in attenuation-corrected myocardial SPECT. J Nucl Cardiol 1999; 6:406. Hudson HM, Larkin RS: Accelerated image reconstruction using ordered subsets of projection data. IEEE Trans Med Imag 13, , Philips Healthcare 166 Astonish Reconstruction NM Application Suite Release 1.0