Vitrea 2. ImageChecker CT (LN-500/LN-1000) version 3.9. User Guide. with VPMC-7861A

Transcription

1 Vitrea 2 version 3.9 with ImageChecker CT (LN-500/LN-1000) User Guide VPMC-7861A

2 Vital Images, Inc Protected by U.S. Patents 5,986,662; 6,130,671; 6,219,059; 7,031,504; Other Patents Pending in the U.S. and other countries. VPMC-7861A Vitrea 2 version 3.9 with ImageChecker CT User Guide (07/2006) No part of this work may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic or mechanical, including photocopying and recording, or by any information storage or retrieval system without permission in writing from Vital Images. Trademarks Vitrea is a registered trademark of Vital Images, Inc. Microsoft, Microsoft Windows, and Microsoft Word are registered trademarks of Microsoft Corporation. ImageChecker is a registered trademark (US) and a trademark (OUS) of R2 Technology, Inc. Restricted Rights Legend If this software or documentation is delivered to the Department of Defense (DOD) of the U.S. Government, it is delivered with Restricted Rights as follows: Use, duplication or disclosure of the software by the U.S. Government is subject to restrictions as set forth in subparagraph (c)(l)(ii) of the Rights in Technical Data and Computer Software clause at DFARS If this software or documentation is delivered to any unit or agency of the U.S. Government other than DOD, it is delivered with Restricted Rights and use, duplication or disclosure by the U.S. Government is subject to the restrictions as set forth in FAR (c)(2). If the software or documentation is delivered to NASA, it is delivered with Restricted Rights subject to the restrictions set forth in (d) of the NASA FAR Supplement. Limits of Liability and Disclaimer of Warranty VITAL IMAGES SHALL HAVE NO LIABILITY OF ANY KIND FOR ANY DIRECT, INCIDENTAL OR CONSEQUENTIAL DAMAGES BASED ON ANY DEFECT, FAILURE OR MALFUNCTION OF THE SOFTWARE,OR USE OF ANY VITAL IMAGES DOCUMENTATION, WHETHER THE CLAIM IS BASED UPON WARRANTY, CONTRACT, TORT OR OTHERWISE. VITAL IMAGES MAKES NO WARRANTY, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO, ANY WARRANTY OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE WHETHER ARISING FROM STATUTE, COMMON LAW, CUSTOM OR OTHERWISE. Notice of Confidentiality This software and the information in this software including, but not limited to, the ideas, concepts and know-how are proprietary, confidential and trade secret to Vital Images, and the information contained therein shall be maintained as proprietary, confidential and trade secret to Vital Images and shall not be copied or reproduced in any form whatsoever. This software and any information contained therein shall not be disclosed to anyone other than authorized representatives of the user's employer, who is contractually obligated not to disclose same without the express written consent of Vital Images. The user of this software and any information contained therein shall not attempt to discern Vital Images' confidential and trade secret information and shall not reverse compile, disassemble, or otherwise reverse engineer this software or any information contained therein. Software License Notice This software is a licensed product of and is distributed by Vital Images, and may only be used according to the terms of that license on the system identified in the License Agreement. In the event of any conflict between these terms and the terms of any written agreement or agreement assented to through electronic means with Vital Images, the terms of such written or assented agreement shall control. MediMark Europe 11 rue Emile ZOLA, BP 2332, GRENOBLE CEDEX 2, France Tel: Fax: medimark@easynet.fr MediMark Europe is an authorized representative in the European Community and acts on behalf of Vital Images, Inc. in the communication of safety-related incidents and regulatory matters with Competent Authorities in the European Community. Distributors are still the first line of communication with their customers regarding service and complaints. VPMC-7861A

3 Table of Contents Overview Safety and Regulatory Considerations Intended Use Warnings and Cautions Warnings: Radiological Interpretation Cautions: System Operation Contact Us Clinical Studies Overview of Clinical Studies CAD Algorithm Description (LN-1000) Available Resources Study Directory Acquiring DICOM Data and CAD results Archiving Volumes with CAD results to CD Studies with CAD Results in the Patient List Loading Studies and Volumes with CAD Results Anonymizing CAD Results Gallery Window Viewer Window Using the Lung Map Using 2D Tools D Viewport Right Mouse Menu Analysis Tab Working with Findings Working with the Measurement Menu Visual Tab Using 3D Tools D Viewport Right Mouse Menu Config Tab Report Window Risk Factors Exporting Text and Images to a DICOM Device Posting Reports to the Intranet Creating Custom MS Word Reports Review Window Setting Up Preferences Exception Handling Communication Errors Unknown Errors Known Errors Required Lung CT Scanning Protocol Table of Contents VPMC-7861A 3 of 64

4 Acquisition Protocol Frequently-Asked Questions of 64 VPMC-7861A Table of Contents

5 Overview Vitrea 2 (Vitrea) is advanced medical imaging software for diagnostic evaluation of computed tomography (CT) and magnetic resonance (MR) image data. Vitrea features real-time navigation of 3D volume data, permitting the user to create two and three-dimensional views of human anatomy and to interactively navigate within these images to better visualize and understand internal structures and disease conditions. R2 Technology, Inc.'s ImageChecker CT (ICCT) software is an advanced image analysis and visualization system designed to assist radiologists in the detection of pulmonary nodules during review of multidetector CT (MDCT) exams of the chest. Vital Images, Inc. has partnered with R2 Technology, Inc. to develop a clinically focused option that seamlessly integrates the ICCT software into the Vitrea software. This separately-licensed Vitrea option allows users to detect lung nodules and visualize studies of the chest from data acquired by a multi-slice CT scanner. There are two Vitrea with ImageChecker CT options: LN-500 A full-featured chest CT review system. If the user finds a suspicious region of interest during study review, they use the Probe tool to determine if the software detected a segmented object at that same location. If the probe finds a segmented object, the user can place a marker on the image. Even if the probe shows that the Vitrea with ICCT software did not segment an object, however, the user can mark it as a finding using the tools on the Vitrea/ICCT Viewer window. LN-1000 A full-featured, lung nodule detection system that uses Computer-Aided Detection (CAD) software to find and mark potential regions of interest (ROIs) for the reader. By default, the system displays the images without CAD markers. After performing the initial read of the exam, the user turns the markers on to review the computer-identified ROIs. IMPORTANT! When working with the Vitrea with ImageChecker CT (LN-1000) option, it is important that you always read unaided by CAD first. Once you evaluate the CAD results, never change your mind about a judgment of actionability of a nodule because CAD fails to mark it. CAD is a tool, and you should always rely primarily on your unaided judgment. The Vitrea with ImageChecker CT (LN-1000) option is designed to be used as an adjunctive reader, providing the radiologist with a second set of eyes with which to find and assess parenchymal densities that may represent clinically significant disease. The ImageChecker CT does this by applying proprietary signal processing CAD algorithms to the large digital datasets generated during scanning. These algorithms analyze the complete set of images and search for findings with features suggestive of a pulmonary nodule. The system conveys information regarding candidate nodules to the workstation for consideration by the reader. It is important to note that the ImageChecker CT CAD System is intended as a back up nodule identification and reporting system for radiologists Overview VPMC-7861A 5 of 64

6 reading all adult MDCT chest exams. This includes everyday diagnostic chest CTs, many of which will have other pathologic processes competing for their attention. After the initial study acquisition by the scanner, two copies of the study are sent out. One copy goes to the Vitrea workstation for you to review and the other copy goes to the server for segmentation processing. When the server completes the segmentation processing, it sends the results to the Vitrea workstation in the form of a DICOM-structured report. When working with the LN-1000 system, the CAD processing results are also sent in the structured report. In the meantime, you can use Vitrea s many visualization tools (for example, cine and 3D volume rendering), to perform an initial read of the study. NOTE Depending on how the system is configured, you can also query a PACS archive and retrieve exams. When this happens, the system treats archived studies as it does new studies. Ask your IT department if you are unsure about whether you can query a PACS archive. When you complete your review, the system makes a summary report available that contains any findings and measurements associated with the study. You can add annotations, arrowtations, measurements, and comments, then print or Intranet-post the report like any other Vitrea report. Safety and Regulatory Considerations PLEASE READ THIS SECTION CAREFULLY BEFORE USING THE VITREA 2 WITH IMAGECHECKER CT OPTION. This section contains information that is essential for the safe and effective use of the Vitrea with ImageChecker CT option. You must understand this information before using this application. For general Vitrea Safety and Regulatory Considerations, refer to the Safety and Regulatory Considerations section of the Basic Vitrea 2 manual orvitrea 2 Essentials manual. Intended Use The Vitrea with ImageChecker CT option is generally intended to be used to assist radiologists in reviewing digital computed tomography (CT) images of the chest. The Vitrea with ImageChecker CT option is specifically intended to be used to display a composite view of 2D cross-sections and 3D volumes of chest CT images, including findings or regions of interest (ROIs) identified by the radiologists or computer-aided detection (CAD) findings. The Vitrea 2 with ImageChecker CT (LN-1000) option is a CAD system designed to assist radiologists in the detection of solid pulmonary nodules during review of multi-detector CT (MDCT) scans of the chest. It is intended to be used as an adjunct, alerting the radiologist after his or her initial reading of the scan to regions of interest that may have been initially overlooked. 6 of 64 VPMC-7861A Safety and Regulatory Considerations

7 This manual is intended to be used by customers who have purchased the separately-licensed Vitrea with ImageChecker CT option. It is written for clinical and administrative staff who use, or are learning to use, the Vitrea with ImageChecker CT option. It assumes a working knowledge of Vitrea and familiarity with the concepts covered in the Basic Vitrea 2 manual or Vitrea 2 Essentials manual. Warnings and Cautions The following Warning and Cautions pertain to the use of the Vitrea with ImageChecker CT option, which contains the ICCT CAD software. CAUTION U.S. Federal law restricts the sale, distribution, and use of this device to or on the order of a physician. Warnings: Radiological Interpretation The CAD software is a detection aid, not an interpretative aid. It should only be used after the first reading of the images. You should always read first before turning on CAD, and then never change your mind about a potential finding if CAD does not mark it. The ICCT CAD software algorithm is designed to detect solid pulmonary nodules that are at least four millimeters and not greater than 30 millimeters in diameter from individual slices in the target scan acquired at a thickness of no less than 0.5 mm and no more than 3.0 mm. The CAD software will not identify all areas that are suspicious for nodules: If the CAD software does not mark a region identified by the radiologist as a nodule, the radiologist should not be dissuaded from working up that finding. The CAD software is not designed to detect changes from other CT exams. The CAD software is more sensitive for the detection of solid spherical nodules than for nodules that are non-solid or part-solid, or have very irregular shapes. The CAD software is not designed to detect ground-glass opacities. The CAD software is not designed to detect other pulmonary parenchymal abnormalities, such as emphysema, atelectasis, consolidation, or diffuse interstitial disease. Scans must be completed in a single breath-hold to avoid motion artifacts. Support for Spiral/Helical scanners with single or dual detectors, and Electron Beam CT (EBCT) is not available at this time. The CAD software marks regions with features suggestive of nodules by overlaying a circle around those regions. These features may not represent nodules, and the user must still use her or his interpretative skills on items marked by the CAD software. Warnings and Cautions VPMC-7861A 7 of 64

8 The CAD software will not enhance what the user sees, rather it helps to identify regions on CT images that should be re-examined. The CAD software will only operate on CT images of the chest that meet the technical requirements outlined in Required Lung CT Scanning Protocol, and the DICOM Conformance Statements for the ImageChecker CT System and Server. Conditions of image quality, such as under-exposure or artifacts (for example, motion artifacts and streak artifacts caused by pacemakers or wires), that diminish radiographic sensitivity may also diminish the sensitivity of the CAD software. Scans grossly distorted due to motion and scans with numerous streak artifacts are likely to be repeated, if possible, since either would affect both human and CAD performance. Individual practice patterns may influence the results obtained when using the CAD software. Therefore, each facility and radiologist should carefully monitor the results that this CAD software has on their practice in order to optimize its effectiveness. Incorrect calibration settings on the monitor can cause details of the image to be lost. Ensure that all users are adequately trained, and that regular preventive maintenance is performed. Cautions: System Operation Operators should review the user guide and receive training as required before using the system. Contact Us For general, non-technical support questions, contact us through our Web site: For customer technical support, contact us using one of the following methods: In the U.S., call the Customer Support line at Outside the U.S., contact your Vital Images distributor. Go to and perform the following steps: a b c In the Destination field, select Technical Questions and Support Issues. Fill out the rest of the form. Click Send. Send an to support@vitalimages.com. To provide feedback about this document or other Vital Images product documentation, send an to feedback@vitalimages.com. 8 of 64 VPMC-7861A Contact Us

9 Clinical Studies This section contains an overview of the clinical studies submitted to the United States Food and Drug Administration (FDA) for the approval of the ICCT System. Overview of Clinical Studies R2 Technology, Inc. has conducted three pivotal clinical studies to evaluate the safety and effectiveness of the ICCT CAD software in the LN The studies were based on a retrospective case collection project involving multiple clinical sites in various regions across the U.S. The clinical studies did not attempt to assess the effectiveness of the device on an asymptomatic lung cancer screening population. The applicability of this device to such a CT screen population has not been established. Objectives The first pivotal study was designed to generate a truth set of cases containing solid pulmonary nodules, as well as cases with no nodules, to be used as a reference truth for subsequent studies. The second pivotal study was an Observer/ROC (Receiver Operating Characteristic) study, designed to measure the performance enhancement of radiologists using the System. The third study was a retrospective study to characterize the stand-alone sensitivity of the CAD system. Sites and Cases for Case Collection Project and Subsequent Studies Five (5) regionally diverse sites contributed 151 cases to the study; two sites in the Northeast, and one site each from the South, the Midwest, and the West. Of these sites, three were private imaging centers and two were academic medical centers (see Table 1). TABLE 1. Sites contributing nodule and non-nodule CT chest cases in the Case Collection Project and subsequent clinical studies Name State Number of Nodulepresent Cases Used in Studies Number of Non-nodule Cases Used in Studies Total Cases Atlantic Medical Imaging NJ MRI & CT Diagnostics VA South Jersey Radiology NJ University of Iowa IA UC, San Francisco CA Total All cases were acquired consecutively from the sites digital archives according to the inclusion and exclusion criteria identified in the case collection protocols. The nodule-present cases collected included only those in which a diagnosis of cancer, either primary lung cancer or an extrathoracic neoplasm, had been documented. Other co-existing disease processes resulting in the formation of nod- Clinical Studies VPMC-7861A 9 of 64

10 ules (e.g., TB, histoplasmosis, rheumatoid lung) were allowed, as were cases containing other underlying pathology such as lobar pneumonia, emphysema, and heart failure. A total of 63 nodule-present cases dating from November 2001 through December 2002 were included in the studies. The study population consisted of 56% females and 44% males, with a median age of 66 (range 20 86). The malignancies consisted of primary lung cancer in 24 (38%) of these cases, and documented extra-thoracic primary cancer with suspected metastatic disease to the lung in the remaining 39 (62%) cases. Forty-six percent (46%) of the exams were performed following injection of intravenous contrast media. The nodule-absent cases collected were those in which no nodules were deemed to be present by the principal investigator at each site. A total of 88 noduleabsent cases dating from June 2002 through December 2002 were included. This group consisted of 53% females and 47% males, with a median age of 55 (range 18 85). Other disease processes could be present, including the presence of pulmonary masses (>3cm). Patients with histories of cancer, radiation therapy, or even previous thoracotomy were allowed. Fifty-two percent (52%) of the exams included intravenous contrast media. STUDY #1 IDENTIFICATION OF REFERENCE TRUTH The objective of this study was to generate a truth set of unanimous actionable nodules, as identified by a panel of three experienced radiologists ( Reference Truth Panel ), to serve as a reference truth for all subsequent studies. To achieve this objective, multiple panel sessions were scheduled in which three radiologists independently read a variable number of cases (min = 12, max = 25) until all 151 study cases had been independently interpreted by three readers. The Reference Truth Panels identified a total of 142 findings in the 151 cases that met the size (4 30 mm) and peak density (greater than 100HU) requirements, and which all three panelists agreed were actionable. We define this set of findings as solid actionable nodules. The presence or absence of at least one of these findings in a quadrant was used as the reference truth for Study #2 below. The findings ranged in size from 4 28 mm. The majority of these findings were between 4 and 8 mm in diameter (46%, 66/142), with the largest categories being the 5 6 mm (15%, 21/142) and 6 7 mm (15%, 22/142) findings. STUDY #2 EFFECT OF CAD SYSTEM ON IMPROVING ACCURACY OF IDENTIFICATION OF ACTIONABLE NODULES The objective of this study was to demonstrate that review of CAD output improves performance of radiologists reviewing MDCT with respect to their ability to accurately identify actionable nodules. The study employed a receiver operating characteristic (ROC) methodology. Ninety (90) cases were randomly selected from the 151 cases in a stratified manner. The cases were divided into four quadrants, yielding 360 regions for evaluation. Each of 15 radiologists independently reviewed the 360 quadrants, first without computer-aided detection (CAD) and then immediately with CAD. The results of this study were used for all of the final analyses of the data. 10 of 64 VPMC-7861A Clinical Studies

11 Each reader rated each quadrant on a actionability scale as to his or her level of confidence that the quadrant contained at least one actionable nodule. Ratings were provided both before and after viewing the CAD marks. For purposes of measuring reader performance, quadrants were defined as actionable if the Reference Truth Panel described in Study #1 unanimously agreed that at least one of the findings in the quadrant was: (1) a solid lung nodule and (2) actionable (i.e., required intervention or short-term follow-up). Otherwise, the quadrant was defined as non-actionable for the purpose of ROC analysis. Before describing the statistical analysis, one example case is described. Figure 1 shows one case where the CAD software pointed out a nodule that was initially missed by four radiologists. The patient was a 67 year-old male, with a history of bladder cancer. A small 4.4 mm noncalcified nodule was present in the right lung, in the lower quadrant, marked by the CAD algorithm. Only eleven of the 15 radiologists rated that quadrant as containing an actionable nodule in their initial review - whereas three more radiologists increased their ratings after viewing the CAD marks. An area under the ROC curve (AUC) was computed for each of the 15 radiologists before and after CAD. The average curve of all 15 radiologists is shown in Figure 2. The area under the ROC curve increased with the use of CAD. If the full plot is viewed as a unit square, the area separating the two curves is Clinical Studies VPMC-7861A 11 of 64

12 FIGURE 1. Example case of a missed nodule where CAD reduced misses 12 of 64 VPMC-7861A Clinical Studies

13 FIGURE 2. Average ROC curve showing pre-cad (dashed line) and post-cad (solid line) performance for 90-case (360 quadrant) study The primary analysis of statistical significance was based on the Dorfman-Berbaum-Metz (DBM) ANOVA-after-jackknife approach 1 (adapted for the fact that the quadrants are clustered data) and the results are presented in Table 2 below. The average reader improvement in AUC (estimated using the ANOVA-after-jackknife) was ± (p = ) with a 95% confidence interval of (0.0084, ). Thus, the study showed a statistically significant improvement in the area under the ROC curve with the use of CAD. Secondary analyses of statistical significance were conducted to determine the dependence of the study results on the use of the consensus reference truth from Study #1 for ROC evaluation. This was done since a reference truth based on a consensus panel assessment of actionability is weaker than one based on biopsy. The first of these variations in the reference truth involved using the findings identified by at least 2 of the 3 members of the Reference Truth Panel in Study #1 (n=310 findings). The ANOVA-after-jackknife analysis was recomputed, with the results as shown below in Table 2. The average reader improvement in AUC was again statistically significant. To further examine the effect of variability in the unanimous three-panelist Reference Truth Panel, three-, two- and single-panelist reference truths were constructed from the data collected in Study #1. Implementing this variable truth is difficult within the framework of the ANOVA-after-jackknife analysis; therefore, the secondary analysis employs a bootstrap analysis. 2 The bootstrap is a computationally-intensive non-parametric method that allows complex analyses to be repeated many times using different randomly generated datasets (all based on 1. Dorfmann DD, Berbaum KS and Metz, CE. Receiver Operating Characteristic Rating Analysis. Invest Radiol 1992; 27: Rutter, C. Bootstrap Estimation of Diagnostic Accuracy with Patient-clustered Data. Acad Radiol 2000; 7: Clinical Studies VPMC-7861A 13 of 64

14 the original data) to approximate the variability that would occur if the entire study was repeated many times. As a test of the validity of the bootstrap mechanism, the analysis was performed first using the unanimous reference truth. As shown in Table 2, the results of the primary ANOVA-after-jackknife analysis and the bootstrap analysis using the unanimous reference truth are very similar. TABLE 2. Study analyses of the significance of the improvement in area under the ROC curve when reference truth or statistical methods were varied Analysis Method Estimated Improvement in AUC p-value 95% CI Primary: ANOVA-after-jacknife, with unanimous reference truth Secondary: ANOVA-after-jacknife, with 2/3 majority reference truth (0.0084, ) (0.0097, ) Secondary: Bootstrap with unanimous reference truth <0.001 (0.0089, ) Secondary: Bootstrap with random 3-panel reference truth <0.001 (0.0076, ) Secondary: Bootstrap with random 2-panel reference truth (0.0077, ) Secondary: Bootstrap with random 1-panel reference truth <0.001 (0.0080, ) Several approaches were used, based on the bootstrap resampling approach, to incorporate random reference truths for the random cases against which the random readers performance could be estimated. Based on varying the reference truth in this way, the average reader improvement in AUC (estimated using the 1000 bootstrap samples with variability in the reference truth) ranged from , as shown in Table 2. Thus again, the results with the varied random reference truth demonstrate that the study showed a statistically significant improvement in the area under the ROC curve with the use of CAD. An additional secondary analysis was performed to determine if the effect of the presence or absence of intravenous contrast in the cases affected the overall study results. The ROC primary ANOVA-after-jackknife analysis (based on the unanimous Reference Truth Panel findings) was redone, stratifying the 90 cases into those that included contrast and those that did not. Analysis of both the 45 non-contrast and 45 contrast media cases showed improvement with CAD (delta AUC for each set of cases was greater than 0.02). Finally, an analysis at the patient level, rather than at the quadrant level, was conducted to determine if there was evidence to show a positive effect of the use of CAD using a methodology closer to the clinical use of the product. Each case in the study was assigned to one case cohort based on what truth nodules were present (depending on the definition of the truth reference standard). For example, the cases that contained at least one unanimous actionable nodule were placed in a case cohort that would be considered true nodule-present cases. R2 also assigned each case a rating of actionable if at least one quadrant had a 14 of 64 VPMC-7861A Clinical Studies

15 rating of 50 or greater (on our scale). A threshold of 50 was chosen based on the way the readers were instructed to use the scale (a rating of 50 was indeterminate ) and the actual ratings data showed a bimodal distribution, with most ratings clearly above 50 or below 50. Thus we can calculate the effect of CAD at the patient level in terms of reduction in observational oversights (i.e., the number of times a patient had a rating change from below 50 pre-cad to above 50 post-cad in a case that contained a truth nodule) and change in the false-positive rate (the same ratings change for a patient with no nodules). This analysis showed that, at the case level, there were 22 corrections of 86 oversights (25.6%) and an increase in the false-positive rate (8 increases in 57 cases; 14.0%). In summary, the primary analysis of the ICCT System study shows a statistically significant improvement in the AUC for an ROC analysis for detecting solid pulmonary nodules between 4 and 30 mm in diameter. This result is robust when different reference truth definitions are used in the analysis. STUDY #3 MEASUREMENT OF CAD ALGORITHM PERFORMANCE The objective of this study was to test that the ICCT System can mark solid pulmonary nodules equal to or greater than 4 mm in size with a high level of sensitivity and a low number of false marks per normal case. The CAD software is designed to specifically identify solid, spherical nodules whereas radiologists identify many other actionable findings (e.g., ground-glass opacities), some of which were included in the set of 142 nodules from Study #1. To evaluate the performance, all 142 unanimous actionable solid nodules arising from Study #1 were first independently shown to a new panel of five radiologists (the Nodule Classification Panel ), each with a minimum of six months experience reading MDCT of the chest. The radiologists were asked to categorize each finding according to its three-dimensional appearance, and determine how closely that appearance approximated a classic pulmonary nodule. The definition of nodule was taken from a standard text 3, and the term classic was as pictured in the same text 4. The descriptor solid referred to a nodule with a peak density greater than 100HU. After all 142 reference truth nodules were reviewed by the Nodule Classification Panel, each candidate nodule was then categorized based on how many of the five Nodule Classification Panel radiologists rated it as classic. The subset of pulmonary nodules deemed to be classic in appearance ( classic nodules ) by at least 4 of the 5 readers (n = 64) was defined as the primary target population for testing purposes. All 151 cases were analyzed by the ICCT System, and an automatic scoring tool matched the CAD marks with the locations of the nodules. The CAD sensitivity results are shown in Table 3. The accompanying false marker rate, as measured on cases with no reference truth nodules, was a median of two marks per case. 3. Fraser RS, Muller NL, Colman N, Paré PD, Eds. Fraser and Paré s Diagnosis of Diseases of the Chest, Fourth Edition, Volume 1. W.B. Saunders, Philadelphia, 1999: page xxxv. The definition is: Round opacity, at least moderately well marginated and no greater than 3 cm in maximum diameter. 4. ibid, Figure 18-31, page 458. Clinical Studies VPMC-7861A 15 of 64

16 TABLE 3. CAD Standalone sensitivity numbers for nodules based on classic ratings Nodule Classification Panel members rating nodule as classic # of nodules CAD Sensitivity* 95% Lower Limit 95% Upper Limit 4/5, 5/5 classic 64 83% 73% 92% 3/ % 44% 94% 0/5, 1/5, 2/ % 21% 44% *CAD Sensitivity is defined as the number of nodules marked by CAD divided by the total number of nodules in each category. A further analysis of the CAD sensitivity for cases with and without intravenous contrast media showed consistent results in the cases with contrast compared to the cases without contrast. The sensitivity for the classic nodules in cases with contrast was 94% (16/17, 95% CI %) and for the classic nodules without contrast was 79% (37/47, 95% CI 67 90%). The sensitivity for all solid nodules (classic and non-classic) in cases with intravenous contrast media is 55% (34/62, 95% CI 43 67%) and in cases without contrast was 61% (49/80, 95% CI 51 72%). The false marker rate measured on the nodule-absent cases was the same in both groups and the same as the summary numbers above. These results confirm that the ICCT System performs at a high level of sensitivity for classic pulmonary nodules, with low false marker rates per normal case. CAD Algorithm Description (LN-1000) The CAD algorithm used by the ICCT server is designed to assist radiologists in the detection of pulmonary nodules during review of multi-detector (MDCT) examinations of the chest, by looking for and identifying solid, parenchymal nodules in the lung tissue. The algorithm also identifies pleural-based nodules, provided they project significantly into the lung. The algorithm generally defines solid, parenchymal nodules as focal densities, approximately spherical in shape, with smooth, lobulated, or spiculated boundaries and with density close to that of soft tissue. The algorithm also recognizes that although pleural-based nodules are not, in general, spherical, their shape usually approximates that of a portion of a sphere. The algorithm does not look for non-nodular abnormalities (e.g., scars, atalectesis, bronchiectasis), nodules or other abnormalities outside of the lung parenchyma (e.g., in the mediastinum or in the chest wall), or ground-glass opacities or other non-solid or part-solid nodules that may represent malignancies. Shape analysis is almost exclusively performed in three dimensions. This means that an object that appears very circular in a particular 2D axial slice may not be marked, because in 3D it is very flat (i.e, sheetlike) or elongated (i.e., linear or tubelike). For instance, on any given slice, a nodule may appear to be a blood vessel, but by considering all of the slices in the stack as a coherent 3D volume, the algorithm is 16 of 64 VPMC-7861A CAD Algorithm Description (LN-1000)

17 able to distinguish a blood vessel, which may be round in one slice but is long and thin in 3D, from a solitary nodule, which is round and compact in every direction. The algorithm can measure and report the volume of any nodule it finds. It can also differentiate between nodules and blood vessels based on density, size, texture, and other properties. This is a task that otherwise can be very time consuming for a radiologist. Based on each finding s size, shape, density, and other characteristics, the algorithm assigns the finding a likelihood. All findings whose likelihood exceeds a fixed threshold are presented to the user as markers. Because the algorithm is tuned to detect nodules that are at least four millimeters in diameter 5, the individual slices in the target scan must be acquired at a thickness of no less than 0.5 mm and no more than 3.0 mm. Slice thicknesses outside of this range can cause unpredictable results. In addition, depending on anatomical variation, image noise, and (rarely) acquisition artifacts (e.g., patient or respiratory motion or streaks caused by very dense objects), a normal anatomical feature may be marked by the algorithm. Pulmonary vessels that appear to the algorithm to change in width may be marked, as may pleural tags or fissure roots that can masquerade as pleural-based nodules. For more details about the CAD algorithm, contact Vital Images, Inc. Available Resources When working with Vitrea with ImageChecker CT, the following resources are available to you: User documentation - This document describes how to use the Vitrea with ImageChecker CT option. Training - The Vital Images Clinical Applications team is available to train your staff if you feel they need additional training on the Vitrea with ImageChecker CT option. Contact Vital Images if you want personalized instruction. Vital Images Customer Support - See the Contact Us section of this manual. Algorithm Documentation - For a general description of how the CAD algorithm processes CT images, see CAD Algorithm Description (LN- 1000) in this manual. DICOM Conformance Statements - On the R2 Technology website ( you can find the DICOM Conformance Statements, which describe the ImageChecker CT implementation of the DICOM protocol. 5. For some studies, it is possible that a nodule with a diameter of less than 4 mm will be found by CAD. Available Resources VPMC-7861A 17 of 64

18 Alert & Update Notification - We are committed to making our products safe and easy to use. If there is an issue that arises or an update with new features is available, we will notify you. 18 of 64 VPMC-7861A Available Resources

19 Study Directory Unlike all other DICOM studies listed on the Study Directory, to process DICOM studies using the Vitrea with ImageChecker CT option, the DICOM data must pass through the CAD server. When processing is complete, the CAD server automatically sends the CAD results to the Vitrea. When the CAD results are received, they are displayed along with their associated volume in the patient list. The line for the CAD results is listed under the associated study, along with the associated volume. The CAD results line starts with a target icon instead of a cube. Acquiring DICOM Data and CAD results You can acquire DICOM data for use with the Vitrea with ICCT option in the following ways: Automatically receiving both DICOM data and CAD results Automatically receiving DICOM data, exporting DICOM data to the CAD server, then receiving CAD results NOTE Do not send more than four datasets to the CAD server at one time. Querying a DICOM device, retrieving DICOM data, exporting DICOM data to the CAD server, then receiving CAD results Retrieving DICOM data from a CD, exporting DICOM data to the CAD server, then receiving CAD results To acquire data exported to both Vitrea and the CAD server: NOTE This happens automatically. The scanner or PACS exports the DICOM data to both Vitrea and the CAD server. The CAD server processes DICOM data, generates results, and exports the results to Vitrea. Vitrea receives both DICOM data and its associated CAD results. Monitor the patient list in the Study Directory for studies listed as both CT and CAD modalities. To acquire data pushed to the Vitrea only: 1 In the patient list on the Study Directory, locate the lung CT study you want to process using the ICCT option. 2 To export the DICOM data to the CAD server for processing, perform the following tasks: If the Study Description contains the word Lung, and the Modality is CT, right-click the volume, and select Export to CAD server... NOTE If the Description in the patient list does not contain the word Lung, the Export to CAD server... menu option will not be visible. Study Directory VPMC-7861A 19 of 64

20 FIGURE 3. Export to CAD server... a b OR If the Export to CAD server... menu item is not available, right-click the volume, then select Export. The DICOM Export dialog box displays. Select the CAD server from list of DICOM devices. Click Export. The CAD server processes DICOM data, generates results, and exports the results to Vitrea. 3 Monitor the patient list in the Study Directory for studies listed as both CT and CAD modalities. To acquire data by querying another Vitrea or DICOM device server: 1 To query another Vitrea workstation, select the tab for the other Vitrea you want to query. NOTE For volumes with CAD results, you cannot restore the workflow from one Vitrea onto another. You can export the dataset from one Vitrea and retrieve it onto another Vitrea, but the second Vitrea will load it as a new study. OR To query a DICOM device server, perform the following steps: a b To narrow the list of results and decrease the time it takes to run the query, enter or specify criteria in any or all of the fields in the DICOM Query area on the Study Directory. In the Server field, click the dropdown arrow. The dropdown menu displays. 20 of 64 VPMC-7861A Study Directory

21 c d Select the server you want to query. Vitrea 2 version 3.9 with ImageChecker CT User Guide Click Query. The system queries the specified device or server and displays a list of all studies and volumes matching the query criteria in the DICOM/CD query list. 2 In the DICOM/CD query list on the Study Directory, right-click the study or volume you want to retrieve. The right mouse button menu displays. FIGURE 4. Retrieve to CAD server and Vitrea 3 If possible, select Retrieve to CAD server and Vitrea. Vitrea sends the query to the DICOM device, and requests that the device sends the DICOM data to the CAD server at the same time. NOTE If the Description in the DICOM/CD query list is anything other than Lung CT, the Retrieve to CAD server and Vitrea menu option will not be visible. OR If the Retrieve to CAD server and Vitrea menu option is not available, perform the following tasks to retrieve both the study or volume and the CAD results: a b c d Select Retrieve to retrieve the study or volume as you would any other study or volume. In the patient list, right-click the retrieved study or volume, then select Export. The DICOM Export dialog box displays. Select the CAD server from list of DICOM devices. Click Export. The CAD server processes DICOM data, generates results, and pushes the results to Vitrea. 4 Monitor the patient list in the Study Directory for studies listed as both CT and CAD modalities. Archiving Volumes with CAD results to CD You can archive Vitrea volumes with CAD results just like all other Vitrea volumes, but to ensure the CAD results are saved along with the volume(s), you must save DICOM files, volume files, and media files. Study Directory VPMC-7861A 21 of 64

22 FIGURE 5. Archive to CD dialog box To archive volumes with CAD results to CD: 1 On the Study Directory, select the volume you want to archive from the patient list. The CAD results are automatically selected along with the volume. The Archive CD dialog box displays. 2 To archive existing reports for this study, check the Reports checkbox. 3 In the Content area, make sure the following checkboxes are also checked: DICOM files Volume files Media files 4 Insert a recordable CD. NOTE If the CD does not have enough free space, or is not a recordable CD, the Record button will be unavailable. 5 Click Record. The progress bar at the bottom of the dialog box updates to show the recording status. When the data has finished recording, the CD tray automatically opens. 6 Remove the CD from the tray. 7 Click Close. To acquire CAD results from CD: 1 Drop the study from the CD into the Vitrea dropbox. 2 Retrieve the study into Vitrea. 22 of 64 VPMC-7861A Study Directory

23 3 Monitor the patient list in the Study Directory for studies listed as both CT and CAD modalities. Studies with CAD Results in the Patient List Studies with CAD results display in the patient list along with their associated Vitrea CT volumes. Studies containing CAD results display as CT CAD modality, because the volumes are listed as CT modality, and their ICCT result data is listed as CAD modality. If a volume has associated CAD results, the results display as an additional line in the patient list. The CAD results icon is a black target instead of a folder (see Table 1). While Vitrea is receiving CAD results from the CAD server, the CAD results target icon includes a progress bar. The CAD results and the volume they are associated with are linked, so if you load the volume, you load the CAD results. When you select a volume or its associated CAD results, Vitrea highlights both lines. You can lock ICCT studies, volumes, and CAD results. No matter which line you lock - the volume or the CAD results, both lines are locked to prevent deletion. When locked, the CAD results status icon changes to red and black with a padlock. TABLE 4. CAD results status icons In process CAD results CAD results error CAD results locked Loading Studies and Volumes with CAD Results You load studies and volumes with CAD results from the Study Directory window just like any other Vitrea study or volume. However, you cannot view multiple volumes with the ICCT preset. NOTE You can load multiple volumes with CAD results into the Gallery window, but when you try to select the ICCT preset, a dialog box displays, saying you must return to the Study Directory and load a single volume. FIGURE 6. Patient list - CT volume and ICCT data Study Directory VPMC-7861A 23 of 64

24 To load a study containing a single volume with associated CAD results: 1 Select the study. 2 Click Load Volume. To load a single volume with associated CAD results: Double-click the volume. OR 1 Select the volume. 2 Click Load Volume. Vitrea loads both the volume and any associated ICCT data and displays the Gallery window. Anonymizing CAD Results You cannot directly anonymize CAD results. You must anonymize the DICOM images, then send the images to the CAD server for processing. To anonymize a case after processing CAD results: 1 Anonymize the DICOM image slices. 2 Resend the anonymized case to the CAD server. 3 Review the anonymized case. To anonymize a case before processing CAD results: 1 Anonymize the DICOM image slices. 2 Send both the anonymized and original slices to the CAD server. 3 Review the original or anonymized case. 24 of 64 VPMC-7861A Study Directory

25 Gallery Window If the study or volume Description in the patient list on the Study Directory is Lung, the pre-assigned protocol on the Gallery window is Lung CT. If this protocol is not selected by default, you should select the Lung CT protocol and the R2 ImageChecker CT preset. This combination of protocol and preset enables the R2 ImageChecker software in the Viewer window. FIGURE 7. Gallery Window - ImageChecker CT preset To select the protocol: In the protocol list, if necessary, select Lung CT. To select the preset: Click Pick for the R2 ImageChecker CT preset. OR Double-click the preset. The Viewer window displays and the volume loads. NOTE The lung map in the upper left corner of the Viewer window takes a few seconds to load. While you are waiting, you can use the 2D tools to start working with the 2D image. Gallery Window VPMC-7861A 25 of 64

26 Viewer Window When you load a lung CT volume using the ICCT protocol into Viewer window, the ICCT software displays three viewports by default. The three viewports contain the following views of the first slice of the loaded volume: 2D lung map (upper left viewport) Initially, a blank 3D viewport (lower left viewport), which will contain a 3D volume when you double-click an ROI in the 2D viewport 2D viewport (large viewport on right) NOTE The following figure is a screen from the LN The R2 button and both the CAD and Discard findings indicators in the Finding Display/Navigation area are only available on the LN FIGURE 8. ICCT/Vitrea Viewer window 26 of 64 VPMC-7861A Viewer Window

27 TABLE 5. Viewer Window features Vitrea 2 version 3.9 with ImageChecker CT User Guide Feature Description Changes the Viewer window format to display 3 viewports (default). Changes the Viewer window format to display 6 viewports. Lung Map Current image indicator line 2D viewport 3D viewport The upper left viewport. Displays a full image of the lungs and any findings marked. Yellow line or rectangle in the lung map. A line identifies the location of the current 2D slice. A rectangle identifies a range of slices, if you specify a slice thickness. In 3 Up format, the right viewport. In 6 Up format, the four right viewports. Displays 2D images. The lower left viewport. Displays a 3D image at the current point of interest (POI). NOTE You must click an ROI in the 2D viewport or lung map before a 3D volume displays in the 3D viewport. Using the Lung Map You use the lung map to see where the current 2D image is located, relative to the anatomy. The location of the current 2D image is represented by the yellow current image indicator line. If you have probed findings, you can use the lung map to see the position of all findings in the study. In the lung map, you can click to navigate directly to an ROI, or drag the current image indicator line, which automatically scrolls through images in the 2D viewport, producing a virtual animation, or melting effect. You can also adjust the window and level settings in the lung map. Viewer Window VPMC-7861A 27 of 64

28 FIGURE 9. Lung map and current image indicator line To navigate directly to an ROI in the lung map: In the lung map, double-click an ROI. OR Click Next or Prev. The system automatically moves the current image indicator line and displays the current slice in the 2D viewport. To melt through the study: Click and drag the current image indicator line up or down the lung map. In the 2D viewport, the application displays each slice as you progress. This creates a virtual animation, or melting effect. NOTE When using this method to navigate through the images, the application does not automatically update the 3D viewport. To update the 3D viewport, double-click the point of interest in the current 2D image. 28 of 64 VPMC-7861A Viewer Window

29 To adjust the window/level settings: 1 Right-click in the lung map viewport, then select Window. The cursor changes to the window/level icon. 2 Click and drag vertically to adjust the window setting, horizontally to adjust the level setting, or diagonally to adjust both. The window and level settings in the lung map are updated. 3 To switch back to navigation mode, right-click, then select Navigate. The cursor changes to crosshairs. Using 2D Tools Both the Visual and Analysis tabs of the Viewer window contain 2D tools for working with the images in the 2D viewport. You can use these buttons to work with the image in the 2D viewport while you are waiting for the segmentation data to load. FIGURE 10. 2D Tools Button Name Use Win/Lev NOTE Alternatively to using the 2D Tool buttons, you can right-click in the 2D viewport to access all of these tools. FIGURE 11. 2D Tools Click to activate the Win/Lev tool. Click and drag on the image to adjust window and level settings. Pan Click to activate the Pan tool. Click and drag the image around in the viewport. Zoom Click to activate the Zoom tool. Click and drag up across the image to minify and down to magnify. Viewer Window VPMC-7861A 29 of 64

30 Button Name Use Measurement Menu (Length tool) Density FIGURE 11 (cont.) 2D Tools Click the dropdown arrow to display the menu. Select from the menu to change the current measurement tool. The button label matches the current tool. Click the button to activate the current tool. The current tool defaults to Ruler (Length) tool. When the Ruler (Length) tool is active, click and drag in the 2D viewport to measure diameter or any straight line. When this measurement tool is active, click anywhere in the 2D viewport to display pixel density (HU). Irregular ROI Elliptical ROI Probe When this measurement tool is active, click and drag in the 2D viewport to trace the border of an irregularly shaped ROI and display all of its measurements except volume. NOTE You can add an irregularly shaped ROI as a 2D finding, and calculate all of its measurements except volume. When this measurement tool is active, click and drag in the 2D viewport to trace the border of an elliptical ROI and display all of its measurements except volume. NOTE You can add an oval-shaped ROI as a 2D finding, and calculate all of its measurements except volume. When this measurement tool is active, click an ROI in the 2D viewport to automatically define and trace the border of the ROI through multiple slices in the volume. To probe an ROI successfully, the CAD server must have recognized it as an ROI. When the probe succeeds, the system automatically displays all measurements, including volume. You can convert a probed ROI to a finding and display all measurements, including volume. To scroll through the 2D image to locate an ROI: In the 2D viewport, move the mouse wheel back and forth. The 2D viewport displays one slice at a time. OR NOTE You can add a probed ROI as a finding. If the probe does not find an ROI, a message displays in the status line. You can still trace the ROI border, add it as a 2D finding, and display its density, area, and other 2D measurements, but the system will not define the finding through multiple slices or calculate its volume. Press the UP ARROW or DOWN ARROW keys on the keyboard. 30 of 64 VPMC-7861A Viewer Window

31 The 2D viewport displays one slice at a time. OR 1 Select the Visual tab. 2 To change the default number of images (5 before the current slice through 5 after the current slice), in the Auto-Cine area, click the up or down arrow buttons. 3 Click Play. The 2D viewport cines through the specified number of slices, from top to bottom and back again. 4 To stop cineing, click Stop. 2D Viewport Right Mouse Menu As an alternative to using the controls and tools on the left side of the Vitrea/ICCT Viewer window, you can right-click in the 2D viewport to display the right mouse menu. The right mouse menu in the 2D viewport contains options for working with findings and for manipulating the image in the 2D viewport, as described below: FIGURE 12. 2D viewport right mouse menu TABLE 6. 2D viewport right mouse menu options Menu Item Use Add finding Alternative to the Add button. See Table 7. Discard/ Delete Delete: For user-defined findings. Alternative to the Delete button. See Table 7. Discard (LN-1000 only): For ICCT findings. Alternative to the Discard button. See Table 7. Prev finding Alternative to the Prev button. See Table 7. Next finding Alternative to the Next button. See Table 7. Left lung Alternative to the Lobe menu. See Figure 20. lobe Right lung Alternative to the Lobe menu. See Figure 20. lobe Viewer Window VPMC-7861A 31 of 64

32 TABLE 6 (cont.) 2D viewport right mouse menu options Menu Item Use Window Alternative to the 2D Win/Lev button. See Table 11. Pan Alternative to the 2D Pan button. See Table 11. Zoom Alternative to the 2D Zoom button. See Table 11. Measure Alternative to the 2D Measurement menu Ruler button. See Table 11. Cine on Alternative to the Play and Stop buttons. See Table 8. Analysis Tab The Viewer window opens to the Analysis tab. On the Analysis tab, you can perform the following tasks: Use the 2D tools and the Measurement Menu Probe 3D ROIs or mark 2D ROIs Convert ROIs to findings Enter comments on findings Indicate the lobe where the finding is located Use buttons to move through all slices with marked findings Display findings by size The number of marked findings is displayed at the top of the Analysis tab. TABLE 7. Analysis Tab Tools Icon Name Use R2 Button Displays all findings identified by the ICCT CAD server that you did not already mark as user-defined. The number of CAD findings indicator displays the total number of findings contained in the CAD results. (LN-1000 only) You can click this button before marking findings on your own, or after, to see if the CAD server identified anything you did not find. (LN-1000 only) CAD Findings Indicator If you have already added a finding matching one contained in the CAD results, it stays marked as user-defined. If you click this button before you mark any findings, you cannot convert CAD findings into user-defined findings. When you display CAD findings, they are outlined by a circular green line in the 2D and Lung Map viewports. If you have already marked a finding contained in the CAD results, it is shown as user-defined, not CAD. CAD findings are only those not first marked by you, but found by the ICCT CAD server. 32 of 64 VPMC-7861A Viewer Window

33 Icon Name Use (LN-1000 only) TABLE 7 (cont.) Analysis Tab Tools Discard CAD findings Indicator User-Defined Findings Indicator Add Vitrea 2 version 3.9 with ImageChecker CT User Guide When you delete a CAD finding, it is only discarded or hidden, not permanently deleted. Discarded findings are outlined by a circular orange dotted line in the 2D and lung map viewports. You can restore discarded CAD findings at any time. The number of discarded findings increments by one each time you discard one, and decrements by one each time you restore one. When you convert ROIs you marked to user-defined findings, their borders are outlined by a green hexagon in the 2D and lung map viewports. You can permanently delete a user-defined finding at any time. The number of findings increments by one each time you add one, and decrements by one each time you delete one. Convert an ROI to a finding. Delete Delete a finding. Prev Display the previous marked finding. Next Display the next marked finding. Working with Findings When you send a volume to the CAD server, the ICCT software detects nodules within the volume, then sends the findings back to Vitrea in the form of CAD results. On the LN-1000, you can access the CAD results by clicking the R2 button to display all CAD findings. For both the LN-1000 and LN-500, you can manually probe ROIs. On the LN-1000, you can use the CAD findings to confirm your manually marked findings, or as an aid in identifying potential nodules. With the LN- 500, you confirm the ICCT results by converting ROIs into findings. The CAD results contain lung location and volumetric measurements. If you detect a finding included in the CAD results, the system will define the finding throughout all images containing it, then automatically display the finding's measurements, including volume. If you detect a finding not included in the CAD results, you can still mark a one-slice (2D) ROI, and convert it to a finding, but the system will not display its volume. The system will display one-slice (2D) measurements only. Viewer Window VPMC-7861A 33 of 64

34 To probe a 3D ROI: NOTE You can convert any probed ROI to a finding. Probed ROIs will convert to 3D volumetric findings. 1 Select the Analysis tab. 2 If the Measurement button is labeled Probe, click it to activate the Probe tool. OR If the Measurement button is NOT labeled Probe, click the dropdown arrow, then select Probe to activate the tool. 3 In the 2D viewport, click the ROI. If the CAD results contain the ROI, the system draws contour lines on every image in the volume that is part of the segmented finding, and re-renders the volume in the 3D viewport with its center on the probed ROI. The status line displays 3D object created by probe. If the CAD results do not contain the ROI, the status line displays No object found by probe. Use the following procedure to mark the ROI as a 2D ROI. NOTE The following figure is a screen from the LN The R2 button and both the CAD and Discard findings indicators in the Finding Display/Navigation area are only available on the LN of 64 VPMC-7861A Viewer Window

35 FIGURE 13. Viewer window, probed finding To mark a 2D ROI: NOTE You can convert any marked 2D ROI to a 2D finding (instead of a 3D volumetric finding). 1 Select the Analysis tab. 2 If the border of the ROI is relatively oval shaped, and the Measurement button is labeled Elliptical ROI, click it to activate the Elliptical ROI tool. OR If the border of the ROI is relatively oval shaped, and the Measurement button is NOT labeled Elliptical ROI, click the dropdown arrow, then select Elliptical ROI to activate the tool. OR If the border of the ROI is irregularly shaped, and the Measurement button is labeled Irregular ROI, click it to activate the Irregular ROI tool. OR If the border of the ROI is irregularly shaped, and the Measurement button is NOT labeled Irregular ROI, click the dropdown arrow, then select Irregular ROI to activate the tool. Viewer Window VPMC-7861A 35 of 64

36 3 In the 2D viewport, click the ROI border and drag to draw an oval or free form line around the ROI. 4 Release the button when you are done tracing the border of the ROI. The system displays the border of the ROI and area (mm^2) in white. NOTE The following figure is a screen from the LN The R2 button and both the CAD and Discard findings indicators in the Finding Display/Navigation area are only available on the LN FIGURE 14. Viewer window, 2D finding To delete an ROI: 1 Select the ROI. 2 Press the DEL key on the keyboard. OR In the View/Edit Properties area, click Delete. The ROI is deleted. To convert any ROI to a finding: 1 When an ROI is selected, its border is white. 36 of 64 VPMC-7861A Viewer Window

37 If the border of the ROI is not white, click it to select it. The border changes to white. 2 In the View/Edit Properties area, click Add. The system converts the ROI to a finding and outlines the finding in the 2D viewport and lung map with a green hexagon. In the Findings/Navigation area, the system increases the number of user findings by 1. In the View/Edit Properties area, the system displays the incremental Image number, the 2D diameter, the average diameter, the average density, and the maximum density of the finding. If the CAD results contain the finding, the system also displays the 3D volume of the finding. NOTE If you want to determine the volume of a 2D finding, you can load the study with one of the Basic Vitrea protocols and presets, then draw contour lines on multiple slices and measure the volume. See the Basic Vitrea 2 manual for more information about how to do this. FIGURE 15. View/Edit Properties area To display CAD findings (LN-1000): 1 Click R2. The system displays any findings the ICCT CAD server identified, that you did not already mark as user-defined findings. NOTE You can display (turn on) the CAD findings before you mark userdefined findings, but you cannot convert CAD findings back into userdefined findings once you turn them on. To delete a user-defined finding: 1 Make sure the finding you want to delete is the currently selected finding. Viewer Window VPMC-7861A 37 of 64

38 NOTE If the finding is currently selected, its measurements display in the View/Edit Properties area. 2 If the finding you want to delete is NOT the currently selected finding, click Next or Prev, or scroll to the finding to make it current. 3 Click Delete. The system deletes the finding. To discard a CAD finding (LN-1000): NOTE You cannot delete CAD findings, but you can discard them so they will not appear on the Report page. 1 Make sure the finding you want to discard is the currently selected finding. NOTE If the finding is currently selected, its measurements display in the View/Edit Properties area. 2 If the finding you want to discard is NOT the currently selected finding, click Next or Prev, or scroll to the finding to make it current. 3 Click Delete. The system discards the finding, but it remains in the CAD results. NOTE The following figure is a screen from the LN The R2 button and both the CAD and Discard findings indicators in the Finding Display/Navigation area are only available on the LN of 64 VPMC-7861A Viewer Window

39 FIGURE 16. Discarded CAD Finding To display discarded CAD findings (LN-1000): In the Finding Display/Navigation area, click in the middle of the Discard CAD findings indicator (orange dotted circle). The system places a checkmark in the circle and displays discarded CAD findings in the 2D and lung map viewports. NOTE If you click in the middle of the Discard CAD findings indicator again (remove the checkmark), the system hides the discarded findings. NOTE The following figures are taken from the LN The R2 button and both the CAD and Discard findings indicators in the Finding Display/Navigation area are only available on the LN Viewer Window VPMC-7861A 39 of 64

40 FIGURE 17. Discard CAD Findings indicator checked (LN-1000) FIGURE 18. Displaying Discarded CAD Findings (LN-1000) To restore a discarded CAD finding (LN-1000): 1 Make sure the finding you want to restore is the currently selected finding. NOTE If the finding is currently selected, its measurements display in the View/Edit Properties area. 2 If the finding you want to restore is NOT the currently selected finding, click Next or Prev, or scroll to the finding to make it current. 3 Click Add. To display findings by size: NOTE You can define the sizes determining large and small findings on the Config tab. 1 In the Show field, click the dropdown arrow, then select one of the following: 40 of 64 VPMC-7861A Viewer Window

41 All findings: Display all findings. Small findings: Display findings smaller than the predefined size threshold. Large findings: Display findings larger than the predefined size threshold. FIGURE 19. Show Findings menu To specify the lobe location of a finding: In the Lobe field, click the dropdown arrow, then select Right unknown, Right upper, Right middle, Right lower, Left unknown, Left upper, Left lower. FIGURE 20. Lobe menu To enter comments for a finding: 1 In the 2D viewport, click the finding. 2 Click in the Comments field and type comments. To move forward or backward through all findings: In the Finding Display/Navigation area, click Prev or Next. NOTE The following figure is taken from the LN The R2 button and both the CAD and Discard findings indicators are only available on the LN Viewer Window VPMC-7861A 41 of 64

42 FIGURE 21. Finding Display/Navigation area Working with the Measurement Menu By default, the Measurement menu button in the 2D toolbar is labeled Ruler. Ruler represents the Length tool. Using the measurement menu, you can perform the following tasks: Display densities Measure the length of straight lines Determine the area of a 2D ROI Probe an ROI throughout multiple slices To work with the measurement menu: Click the button to activate the tool represented by the current button label. OR Click the dropdown arrow to display the menu, then select a different tool. FIGURE 22. Measurement menu 42 of 64 VPMC-7861A Viewer Window

43 To display pixel density (HU): 1 If the Measurement button is labeled Density, click it to activate it. OR If the Measurement button is not labeled Density, click the dropdown arrow, then select Density. 2 Click the 2D image where you want to see the pixel density The system displays the pixel density (HU) at the point where you clicked. To measure length: 1 If the Measurement button is labeled Ruler, click it to activate it. OR If the Measurement button is NOT labeled Ruler, click the dropdown arrow, then select Length. 2 Click on a starting point, drag to an end point, then release. The system draws a line and displays its length (mm). Visual Tab On the Visual tab, you can use the same 2D tools as on the Analysis tab to perform the following tasks: Use the 2D tools and the Measurement Menu Measure and probe ROIs NOTE You must switch to the Analysis tab to convert an ROI to a finding. Use 3D tools to manipulate the image in the 3D viewport Cine through a specified number of slices Change window/level visual settings Show or hide the chest wall, borders, markers and measurements, and patient information Adjust slice thickness Viewer Window VPMC-7861A 43 of 64

44 FIGURE 23. Viewer Visual tab TABLE 8. Visual Tab features Feature Name Use Play/Stop When not cineing, button is the Play button. Click to cine through the specified number of images before and after the current slice. When cineing, button is the Stop button. Click to stop cineing. Number of Cine Images Specify the number of images to scroll through. Starting at the current slice, the system scrolls X number of slices above the current slice, and ends X number of slices below the current slice. The system keeps scrolling through the current range of slices until you click Stop. Click the up or down arrow to increase or decrease the number of images to play (cine). 44 of 64 VPMC-7861A Viewer Window

45 TABLE 8 (cont.) Visual Tab features Feature Name Use Window Presets list Click the dropdown arrow and select predefined window and level settings from the dropdown list: Bone, Lung, Mediastinal. The default is Lung. You can set up custom window and level settings on the Config tab. View options Choose from the following options: Show/Hide chest wall in the 3D viewport Show/Hide borders Show/Hide markers/measurements Show/Hide patient information Multi-slice images Display the original slices averaged. Specify the number of slices to display averaged in the 2D viewport(s). The lung map current image indicator line changes to a rectangle, indicating the range of slices. To cine: 1 To change the default number of images (default is 5 before the current slice through 5 after the current slice), in the Auto-Cine area, click the up or down arrows. 2 Click Play. The 2D viewport cines through the specified number of slices, from top to bottom and back again. 3 To stop cineing, click Stop. To change the preset window and level settings: In the Window presets field, click the dropdown arrow, then select Bone, Lung, or Mediastinal. OR If you have set up custom presets, select from the list. To adjust view options: In the View Options area, choose from the following options: Show/Hide chest wall: Displays or removes the chest wall from the volume in the 3D viewport Show/Hide borders: Displays or removes the contour lines around probed ROIs or findings on all slices Viewer Window VPMC-7861A 45 of 64

46 Show/Hide markers/measurements: Displays or removes finding markers and any measurements from the viewports Show/Hide patient information: Displays or removes patient information from the viewports To change slice thickness: 1 In the View Options area, make sure the Multislice images box is cleared. 2 To change the default number of slices (5), click the up or down arrows. Using 3D Tools NOTE The 3D tools are active only after you have selected an ROI in the 2D or lung map viewport and the 3D viewport contains a volume. The Visual tab of the Viewer window contains 3D tools you can use to manipulate the image in the 3D viewport. Using these tools, you can rotate the volume on its current horizontal or vertical axis, magnify or minify the volume, and add or remove tissue surrounding the ROI. NOTE Alternatively to using the 3D Tool buttons, you can right-click in the 3D viewport to access all of these tools. FIGURE 24. 3D Tools TABLE 9. 3D Tools Button Name Use Rotate Click to activate the Rotate tool. In the 3D viewport, click and drag to rotate the volume. Zoom Tissue Click to activate the Zoom tool. In the 3D viewport, click and drag up to minify, or see more of the surrounding area, or down to magnify, or see less of the surrounding area. Click to activate the Tissue tool. In the 3D viewport, click and drag up over the 3D volume to remove low-density soft tissue, or down to restore it. 46 of 64 VPMC-7861A Viewer Window

47 3D Viewport Right Mouse Menu As an alternative to using the buttons and controls on the left side of the Vitrea/ ICCT Viewer window, you can right-click in the viewport to display the right mouse menu. The 3D viewport right mouse menu contains options for adding and deleting findings, and for manipulating the 3D volume, as described below: FIGURE 25. 3D viewport right mouse button menu TABLE 10. 3D viewport right mouse button menu options Menu Item Use Add finding Alternative to the Add button. See Table 7. Discard/Delete Delete: For user-defined findings. Alternative to the Delete button. See Table 6. Discard (LN-1000 only): For ICCT findings. Alternative to the Discard button. See Table 6. Prev finding Alternative to the Prev button. See Table 7. Next finding Alternative to the Next button. See Table 7. Rotate Alternative to the 3D Rotate button. See Table 9. Zoom Alternative to the 3D Zoom button. See Table 9. Tissue contrast Alternative to the 3D Tissue button. See Table 9. Config Tab On the Config tab, you can perform the following tasks: Set up custom Site window/level settings Select the default window/level setting Change the width and level values of predefined window/level settings Specify the large and small size values for displaying findings by size Viewer Window VPMC-7861A 47 of 64

48 FIGURE 26. Window Presets area To create a custom site window/level setting: 1 Select Site. The list of presets displays Window 1, Window 2, Window 3, up to 5 custom presets. 2 Select the preset you want to define. To rename the preset, do the following steps: a b c Click Rename. In the Edit the Preset Name field, type the new name. Click OK. 3 In the Width field, enter a new width setting. 4 In the Level field, enter a new level setting. To change the custom preset choices in the Window/Presets dropdown list on the Visual tab: 1 Select Site. The list of presets displays Window 1, Window 2, Window 3, up to 5 custom presets. 2 To include the custom preset in the Window/Presets dropdown list, check the box in front of the preset. OR To remove the custom preset from the Window/Presets dropdown list, clear the box in front of the preset. 3 Click Apply. The Window/Presets dropdown list on the Visual tab is updated. 48 of 64 VPMC-7861A Viewer Window

49 To change the default window/level setting: 1 To use one of the predefined application settings as the default window/level setting in the Viewer window, select Application. The list of presets displays Bone, Lung, and Mediastinal. OR To use one of your site settings as the default window/level setting in the Viewer window, select Site. The list of presets displays Window 1, Window 2, Window 3, up to 5 custom presets. 2 Select the preset you want to use as the default. 3 Click Default. To rename a site window/level setting: NOTE You cannot rename the predefined Application window/level settings. The Rename button is not available. 1 Select the window/level setting you want to rename. 2 Click Rename. The Rename dialog box displays. 3 In the Edit the Preset Name field, type the new name. 4 Click OK. To change assigned width or level settings: 1 Select the window/level setting name you want to change. 2 To change the width, type a new value in the Width field. 3 To change the level, type a new value in the Level field. 4 Click Apply. To specify the finding size threshold: In the Finding Size Threshold area, click the up or down arrows. FIGURE 27. Finding Size Threshold area Viewer Window VPMC-7861A 49 of 64

50 Report Window For ICCT studies, the Vitrea Report window takes on a predefined, automaticallygenerated format. The first page contains institution, patient, and exam information, as well as a summary of findings and a place to enter comments. At the bottom of the first page, you can specify patient risk factors. ICCT creates a snapshot of every image containing a finding and places it in the slide tray on the Report page. A target icon is used to represent the ICCT snapshots. This is the same target icon as shown in Table 1. You cannot save or restore ICCT workflows from these snapshots, nor can you export them in primary DICOM format. After the first page, a page is automatically generated and added to the report for every snapshot created by ICCT. The auto-generated snapshot summary pages contain the snapshot and information about the specific finding. In addition, you can enter a shape and observations for each finding. You can annotate snapshots on the automatically-generated pages. You cannot delete, reorganize, or replace them. If necessary, you can add additional pages after the automatically-generated pages, and work with them just as you would any other Vitrea report. You can finish an ICCT report like you finish any other Vitrea report. You can send it to a postscript printer, post it to the Intranet, create a Microsoft Word report from it, or archive it to CD. The only thing you cannot do with an ICCT report is export it in true DICOM format. NOTES Contact Vital Images Customer Support for help customizing your Word report template, setting up predefined risk factors and categories, as well as finding shape presets. Additional pages added after the auto-generated pages on the Report tab will not be included in Word reports. Risk Factors You can configure your own risk factors in Vitrea preferences. You can define up to three groups, containing up to four risk factors each. The risk factors will be added to the Word report template. EXAMPLE Patient History: Lung Disease (Yes/No) Cancer (Yes/No) Others Exposure Risk: Live with Smoker (Yes/No) Heavy Metal (Yes/No) Asbestos (Yes/No) Other 50 of 64 VPMC-7861A Report Window

51 Smoking History: Smoker (Yes/No) From Age (type a number) To Age (type a number) Packages/Day (type a number) NOTE Contact Vital Images Customer Support for help setting up predefined risk factors and categories. Auto-generated information: Lung - Displays Right or Left if you specified the lobe location of the finding. Otherwise, displays Unknown. Density (HU) - Displays the voxel density of the finding. Volume (mm3) - If the finding is contained in the CAD results, displays the volume of the finding, as measured by the probe. Diameter (mm) - Displays the average diameter of the finding. Long Axis (mm) - Displays the farthest distance between any two points on the border of the finding, taken from the slice where the finding has the largest area. Short Axis (mm) - Displays the longest perpendicular distance between any two points on the border of the finding, taken from the slice containing the long axis. Lobe - Displays the lobe, if you specified one. Otherwise, displays Not Specified. Slice # - Displays the image where the finding was probed, then the system-assigned finding number. User-supplied information: Shape - Select a shape from the dropdown list or type your own in the field. The default shape is blank. Choose from the following options, or type a different shape description: Spiculated, Stippled, Popcorn, Punctate, Irregular, Lobular, Smooth, or <type other value>. NOTE Contact Vital Images Customer Support for help setting up predefined finding shapes. Observations - Enter comments in the field. Report Window VPMC-7861A 51 of 64

52 FIGURE 28. Report window To enter risk factors: 1 In the Risk Factors area, click the dropdown arrow on any of the risk factor category buttons. A dropdown menu displays a list of risk factor choices. 2 Select one of the choices. NOTE Any risk factors you select on the Report page will be lost if you return to the Viewer window. OR 1 Click in the field. 2 Type a risk factor. OR 1 Click in the field. 2 Press the UP ARROW or DOWN ARROW keys on the keyboard. To enter comments: In the Comments field, type your comments. 52 of 64 VPMC-7861A Report Window

53 NOTE Any comments you enter on the Report page will be lost if you return to the Viewer window. Exporting Text and Images to a DICOM Device Comments, patient, and institution information on the Report tab will not normally export to a DICOM device. You can only export images in true DICOM format. However, you can annotate and add arrows to snapshots on a report page, and then export to a DICOM server. This allows you to add some documentation to the exported images. To export text and images to a DICOM device: 1 On the report page(s), annotate any snapshots you want to export. 2 Add arrows to the snapshots, if necessary. 3 Click Export. The DICOM Export dialog box displays. 4 Select the DICOM server. 5 Click Export. NOTE All ICCT images are automatically annotated, and are exported in DICOM Secondary Capture format. Posting Reports to the Intranet You can review ICCT reports just like all other Vitrea reports. Anyone with access to your organization's Intranet, and the Vitrea report browser can use the Review window. Posting reports to the Intranet allows you some file management capabilities. Once you post a report to the Intranet, it is saved on your workstation, even if you decide to delete the study from the Study Directory in Vitrea. To post a report to the Intranet: On the Report tab, click Intranet Post. Creating Custom MS Word Reports A predefined Word report template is included with the ICCT option. The report template includes all of the information included on the automatically-generated report pages. If you do not specify or enter information about each finding or add comments and observations on the Report pages, you can enter all of that same information into the report in Word. NOTE Contact Vital Images Customer Support for help customizing your Word report template. To create a custom Word report: 1 Click Word Report. Word launches a customizable template, containing all of the information you entered on the auto-generated pages in the Report window. You can change or add information in any field. NOTE Additional pages added after the auto-generated pages on the Report tab will not be included in Word reports. Report Window VPMC-7861A 53 of 64

54 FIGURE 29. Predefined Word report template 54 of 64 VPMC-7861A Report Window

55 If you added details while examining this study, multiple pages can be appended to the title summary page: FIGURE 29 (cont.) Predefined Word report template Report Window VPMC-7861A 55 of 64

56 Review Window You can review ICCT reports just like all other Vitrea reports. Anyone with access to your organization's Intranet, and the Vitrea report browser can view reports from the Review window. FIGURE 30. Review Window browser - Report Index 56 of 64 VPMC-7861A Review Window

57 FIGURE 31. Review Window browser - Reports list Vitrea 2 version 3.9 with ImageChecker CT User Guide To view a Word report: On the Vitrea Report page, click the Word icon in front of the patient name. To view an Intranet report: On the Vitrea Report page, select the patient name. Review Window VPMC-7861A 57 of 64

58 FIGURE 32. Review window browser - Intranet report Setting Up Preferences You can set up risk factors and finding shape options in Vitrea preferences. You can also have Vital Images set up custom Word report templates with your institution's logo and standard text. NOTE For help setting up these preferences, contact Vital Images Customer Support. Exception Handling When using Vitrea with ICCT, you may occasionally receive an error message. ICCT errors fall into the following categories: Communication Errors - The communication link between Vitrea and ICCT has shut down. Unknown Errors - Internal errors for which ICCT cannot determine a cause. 58 of 64 VPMC-7861A Setting Up Preferences

59 Known Errors - Errors caused by inappropriate system conditions or actions taken by the user. Communication Errors If ECF stops running for any reason during the Vitrea session, the following dialog box displays. Restart Vitrea. FIGURE 33. Communication failure dialog box Unknown Errors If the ICCT software experiences an unknown error, Vitrea returns to the Gallery window. A dialog box displays the following message: An error occurred. Lung CAD software will have to shutdown. ICCT:Error Loading.Mandator attribute missing. ICCT:Error Loading. Invalid series type. ICCT:Error Loading. Multiple series images. ICCT:Error Loading. Invalid image file. ICCT:Error Loading. SR SC UID mismatch. ICCT:Error Loading. SR Image UID mismatch. ICCT:Error Loading.Cannot load SR. ICCT:Error Loading ICCT:Error creating scout. ICCT:Error Loading.Mismatch between number of images and CAD results calcuations. If you receive any of these error messages, try doing the following steps: Reselect the ICCT preset to reload the study from the Gallery window. If you still receive an error message, restart Vitrea, then reload the study. If you still receive an error message, reboot the workstation, then reload the study. If you still receive an error message, resend the dataset, then reload the study. If none of the previous steps resolves the error, contact Vital Images Customer Support. Exception Handling VPMC-7861A 59 of 64