HepaTux A Semiautomatic Liver Segmentation System

Size: px
Start display at page:

Download "HepaTux A Semiautomatic Liver Segmentation System"

Transcription

1 HepaTux A Semiautomatic Liver Segmentation System Andreas Beck and Volker Aurich Institut für Informatik, Heinrich-Heine-Universität Düsseldorf, D Düsseldorf becka-miccai@acs.uni-duesseldorf.de Abstract. While the MICCAI challenge strives for fully automatic liver segmentation, we present a fast semiautomatic method that is capable of segmenting livers with the help of an experienced human supervisor within 3 to 15 minutes. As fully automatic methods need to be checked for errors by a human observer anyway, we believe that for many practical applications the presented method will be useful. 1 Introduction HepaTux [1] is a framework for liver segmentation we developed using the EC- CET Toolkit [2] developed at our university. It features several methods to segment liver tissue and vessels within the liver. For the purposes of the challenge we implemented a new wizard called NLC NTX for the current development version of HepaTux. ECCET wizards are interactive web applications that guide the user through the steps of the segmentation process by both giving advice on how to proceed and by controlling the running ECCET program to trigger complex actions. ECCET is capable of loading a variety of file formats including ITK metaheader files as used in the challenge as well as DICOM image stacks and it supports loading DICOM data directly from PACS servers. This makes ECCET based applications easy to integrate into clinical workflow. 2 Method 2.1 Workflow After starting the main application and launching the Hepatux wizard, the user is prompted to load the liver dataset to process. A few options are offered for converting unusual datasets. Then three orthogonal planar views of the dataset are presented and the following steps are performed without any further preprocessing.

2 Step 1: Interactive filling The user clicks at any voxel he considers to be liver tissue and a three-dimensional fill algorithm will start to flood-fill the volume from there. The NLC criterion (described below in detail) serves as a stop condition. It forces the algorithm to stop wherever the distribution of the grey values in the neighbourhood differs significantly from that at the seedpoint. By a control parameter which is not very critical the stop condition can be adapted to the particular noise level of the liver tissue. Usually liver tissue and lesions are distinguished by the user and marked with different colors. For the purpose of the challenge lesions were treated as liver tissue. This filling step is repeated by choosing new seedpoints until the liver is segmented in its entirety. The choice of the seedpoints is intuitive and does not require a high level of accuracy since remaining small gaps (e.g. at large vessels and the border of lesions) will be filled by the postprocessing step. The three-dimensional nature of the algorithm allows to fill the liver quickly. In order to check the progress of filling it is dangerous to use only one planar view because misinterpretation of the local shape of the liver easily occurs. This effect can even be seen in some of the provided reference segmentations (see the rectangular artifacts on the first line of figures 1 and 2). Therefore HepaTux offers instant visualization of the filling state in all three planar views resulting in an intuitive and non-tedious procedure. The time needed for the whole filling step depends on the complexity of the situation. Datasets with very inhomogeneous texture of the liver tissue and irregular lesions like the testing example # 3 in the challenge require more time than those with quite homogeneous textures of liver and lesions. Step 2: Touching up the result in the 3D view During the whole segmentation process a 3D view of the current result is available to visualize progress. After the filling step this view is used to identify and correct remaining problematic parts. Typical problem areas are the apex of the heart, small pieces of tissue between the ribs or parts of the stomach touching the liver. Such errors are easily spotted in the 3D view and corrected using various intuitive tools (like the virtual knife for cutting off unwanted pieces) for 3D volume operations. If the 3D view reveals missing parts, one can easily go back to the fill step. For the purposes of the challenge, the vessels leading to the liver were also cut away using this tool to match the standards set by the training datasets. Note that we did not emulate the training datasets property of including internal vessels 1. Including vessels dependent on a specific view does not seem to be anatomically justified but rather inspired by the reference segmentation methodology. We have run postprocessing filters on our datasets to emulate this property, and improvements are minimal (about 0.1% higher Dice similarity or lower error). The influence on distance based scoring systems would probably be higher, though. 1 From the comments to the reference segmentation: In general, a vessel counts as internal if it is completely surrounded by liver tissue (in the transversal view).

3 Step 3: Postprocessing After the 3D touch up, a fully automated postprocessing step follows. Its purpose is the elimination of the gaps left after the filling step and the smoothing of the 3D shape of the surface of the segmented object. Both is achieved by creating a limited convex hull: The segmentation result is extended by all straight lines with endpoints on border voxels which lie within a fixed size bounding box. The whole postprocessing step takes about one minute to compute. 2.2 Employed algorithms The NLC criterion The breadth first search of the interactive filling step is controlled by a criterion which we call non-linear coupling (NLC). Its definition is similar to that of the non-linear Gaussian filters in [3][4][5] which are used for edge-preserving smoothing. Let f(q) be the grey value of the voxel q and σ and ζ be positive numbers. Then the non-linear coupling to a level L in a voxel p is defined as NLC (σ,ζ,l) f(p) = g σ ( q p ) g ζ ( f(q) L ) q Here g λ (t) = exp( t2 2λ ) is the non-normalized Gaussian function with standard 2 deviation λ > 0, and the sum extends to all voxels q in a rectangular neighbourhood of p with edge length of approximately 6σ. In the above-mentioned filling step the level L is chosen as the linear Gaussian mean of the grey values in the neigbourhood of the selected seedpoint. During the breadth first search the NLC value to this level L is calculated for each new voxel. Only if this value is sufficiently high the voxel is included in the segmented volume and in the search queue for further expansion. The parameter σ is fixed; for usual CT data σ = 2 is well-suited (assuming the distance of adjacent voxels is 1). The parameter ζ is adapted to the liver texture. A good value is the estimated standard deviation σ grey of the grey value distribution of the tissue to be filled. A tool is included to help getting an estimate of σ grey if desired. The choice of this parameter is not particularly critical, so keeping it at the default value will work for most datasets. When chosen too low the filling algorithm stops very soon and has to be restarted more often. To facilitate adding areas with irregular, not very homogeneous texture it is advantageous to start the filling step with a larger ζ assuring the filling of the immediate vicinitiy of the clicked seedpoint and then gradually lowering ζ while progressing in the breadth first search. A good strategy is to choose as initial value ζ 0 = 3σ grey and then to update the current value ζ n according to ζ n = 0.9 ζ n σ grey after each iteration over the breatdh first search queue (i.e. after each layer added). Despite the simplicity of the criterion it has turned out in many experiments to be a reliable condition to stop the filling at texture boundaries. The virtual knife Interaction with 3D objects for the purpose of sculpting is known to be difficult as only few operations are intuitive and efficient. One such operation we employ is a virtual knife. When clicking at a point of the object in the 3D view the object is cut along the vertical plane through the clicked point

4 in viewing direction. This is achieved by flood-filling within a small vicinity of the cut-plane that is defined by the clicked voxel, the camera location and the camera up -vector. At the clicked point a breadth first search is started that will terminate when it deviates too much from this cut-plane. The method provides intuitive behaviour much like cutting with a real knife: It will only go through the first object seen, leaving objects behind untouched. 3 Results 3.1 Performance on the training datasets File Time Vol. Over- Dice Sim. Difference number (in s) lap Error Coefficient Reference HepaTux % 95.5% 5.8% 3.8% % 95.4% 6.0% 3.6% % 94.8% 4.9% 6.0% % 97.3% 2.5% 3.1% % 96.9% 4.0% 2.3% % 96.4% 3.0% 4.5% % 96.8% 3.1% 3.5% % 96.3% 1.9% 5.7% % 96.8% 3.6% 2.9% % 97.2% 3.2% 2.6% % 97.1% 3.3% 2.7% % 97.2% 1.8% 3.9% % 94.2% 3.7% 8.7% % 96.4% 3.2% 4.3% % 97.0% 3.9% 2.2% % 97.0% 2.5% 3.8% % 96.1% 5.0% 3.0% % 96.2% 6.1% 1.8% % 97.0% 4.1% 2.2% % 95.3% 6.6% 3.2% Average: % 96.4% 3.9% 3.7% Worst case: 11.0% 94.2% 6.6% 8.7% Table 1. Performance of HepaTux on the challenge training datasets Table 1 shows the performance of the method on the training datasets made available for the challenge. In this and all other tables, the figures were derived as follows: If A and B are the voxel sets obtained by two segmentations then Volumetric Overlap is calculated as the ratio #(A B) #(A B) and Volumetric Overlap Error is the difference of that value to 1. 2#(A B) Dice Similarity Coefficient is the ratio #(A B)+#(A B). Difference is the ratio #(A\B) #(B\A) #(A B) or #(A B) respectively, given to provide an idea about the symmetry of the errors. If e.g. a volume is entirely contained in the other, its difference value is 0. Segmentation time varied between about 3 and 15 minutes, with an average of 7 minutes. As this already includes the time taken for error-checking and correction of results, we think this method is feasible for clinical use.

5 File Vol. Over- Dice Sim. Difference number lap Error Coefficient 1st 2nd 1 3.6% 98.2% 0.9% 2.8% 2 5.5% 97.2% 4.1% 1.7% 3 6.0% 96.9% 4.4% 2.0% 4 4.1% 97.9% 2.9% 1.3% 5 3.6% 98.2% 1.1% 2.6% 6 3.3% 98.3% 1.4% 2.0% 7 3.0% 98.5% 2.0% 1.1% 8 4.6% 97.7% 4.3% 0.5% 9 4.3% 97.8% 1.0% 3.5% % 97.6% 4.0% 0.9% % 97.6% 1.7% 3.1% % 98.2% 1.7% 1.9% % 97.5% 2.6% 2.5% % 97.4% 3.6% 1.7% % 98.1% 1.9% 1.9% % 96.7% 5.7% 1.0% % 97.3% 3.3% 2.3% % 97.3% 1.5% 4.1% % 97.8% 2.1% 2.3% % 96.3% 5.3% 2.4% Average: 4.6% 97.6% 2.8% 2.1% Worst case: 7.1% 96.3% 5.7% 4.1% Table 2. Intra-Observer variance (training datasets) Intra-/Inter-observer-variance Semiautomatic methods inherently depend on user input. However, for clinical practice it is important to get reliable results independent of the variations in inputs provided by the same user and even more importantly by other users. Therefore, the training measurements were repeated by the same observer (A. Beck) and a second observer (V. Aurich) to estimate the degree of intra- and inter-oberserver-variance. Results of these measurements are shown in Tables 2 and 3. File Vol. Over- Dice Sim. Difference number lap Error Coefficient 1st 2nd 1 5.9% 96.9% 1.9% 4.4% 2 7.5% 96.1% 3.6% 4.5% 3 7.2% 96.3% 4.9% 2.8% 4 4.8% 97.6% 4.1% 0.9% 6 5.0% 97.5% 2.2% 3.0% 7 5.8% 97.0% 5.5% 0.6% 8 4.0% 98.0% 3.0% 1.2% % 97.7% 1.1% 3.7% % 96.0% 2.6% 5.7% % 95.8% 6.3% 2.6% Average: 6.1% 96.9% 3.5% 2.9% Worst case: 8.1% 95.8% 6.3% 5.7% Table 3. Inter-Observer variance (training datasets)

6 3.2 Performance on the test datasets Internal tests The challenge test datasets were also segmented twice to acquire an intra-observer confidence measure. Segmentation time varied between about 4 and 15 minutes, with an average of 7.5 minutes. Table 4 gives the times taken to segment each liver and the intra-observer variance. File Time/s Vol. Over- Dice Sim. Difference number 1 2 lap Error Coefficient % 98,0% 2,7% 1,4% % 97,7% 2,6% 2,0% % 97,3% 2,1% 3,5% % 96,2% 2,2% 5,7% % 96,3% 2,8% 4,8% % 97,3% 1,7% 3,9% % 97,4% 0,7% 4,7% % 97,2% 2,1% 3,6% % 97,9% 2,2% 2,1% % 95,8% 2,1% 6,7% Average: % 97,1% 2,1% 3,8% Worst case: % 95,8% 2,8% 6,7% Table 4. Intra-Observer variance (test datasets) Challenge evaluation Both segmentation series were submitted for evaluation. The results are shown in tables 5 and 6 and figures 1 and 2. 4 Discussion 4.1 Strengths and Weaknesses As readily seen from the middle columns of figures 1 and 2, our method is less prone to creating rectangular-looking artifacts than manual segmentation using plane-based methods. In areas where delineation of the target tissue is not very pronounced (like in areas where the used plane is almost tangential to the target volume) or where exemption of parts would require extra work (like exempting the internal vessels), this can lead to very unnatural-looking results when viewed in another plane. However, the algorithm does not try to create smooth borders, except for a local smoothing step at the end. This yields visually less appealing results near borders where the grey values do not present a clear edge. This can be seen most readily in the bottom row, where a large, structured lesion impedes segmentation. As stated in the results, the authors were also the observers. We thus assume that the required time for an observer that uses the tool only occasionally might

7 Dataset Overlap Error Volume Diff. Avg. Dist. RMS Dist. Max. Dist. Total [%] Score [%] Score [mm] Score [mm] Score [mm] Score Score Average Table 5. Segmentation 1: Results of the comparison metrics and scores for all ten test cases. Fig. 1. Segmentation 1: From left to right, a sagittal, coronal and transversal slice from a relatively easy case (1, top), an average case (4, middle), and a relatively difficult case (3, bottom). The outline of the reference standard segmentation is in red, the outline of the segmentation of the method described in this paper is in blue. Slices are displayed with a window of 400 and a level of 70.

8 Dataset Overlap Error Volume Diff. Avg. Dist. RMS Dist. Max. Dist. Total [%] Score [%] Score [mm] Score [mm] Score [mm] Score Score Average Table 6. Segmentation 2: Results of the comparison metrics and scores for all ten test cases. Fig. 2. Segmentation 2: From left to right, a sagittal, coronal and transversal slice from a relatively easy case (1, top), an average case (4, middle), and a relatively difficult case (3, bottom). The outline of the reference standard segmentation is in red, the outline of the segmentation of the method described in this paper is in blue. Slices are displayed with a window of 400 and a level of 70.

9 be higher. However, there are no complex decisions to be made that require background knowledge of image processing in general or the system in particular. We thus do not expect that usage by a non-expert would cause large timing impacts. We would expect accuracy to increase when the system is used by an expert of the anatomy side. However, the testing results score our segmentation about equal to the student performing the second reference segmentation. We are currently planning to put this version of the software into clinical testing to see how it performs in the hands of radiologists that are not yet proficient with ECCET applications. Some preliminary tests with a radiologist that has been using other applications of the ECCET toolkit showed good acceptance and good segmentation results at acceptable time after a supervised first segmentation. 4.2 Overall Score The scoring system was set up to give a score of 75 for a second human observer. We are thus very pleased to see our system perform at that value for both submitted results on average and no total score going below 66. These results seem acceptable, especially as the segmentation was done by a computer scientist with only rudimentary anatomical knowledge. Due to our ignorance of the internal vessel rule, we were expecting higher deviations in the distance scores, especially the Maximum Distance Score. Looking at absolute values, we see those expected high deviations but they seem to be comparable to those arising from intra-observer differences used to calibrate the scoring system, which is surprising. Comparing the overlap errors we retrieved from our intra-observer testing with those provided by the challenge evaluation, we find the errors to be slightly (1.3% average, 3.2% maximum) higher. This matches our finding from the training datasets, that the method has an about 1.5% higher overlap error on the inter-observer case. References Aurich V., Weule J.: Non-Linear Gaussian Filters Performing Edge Preserving Diffusion. Proceedings 17. DAGM Symposium über Musterekennung, Springer , Aurich V., Mühlhaus E., Grundmann S.: Kantenerhaltende Glättung von Volumendaten bei sehr geringem Signal-Rausch-Verhältnis. Proceedings Bildverarbeitung fr die Medizin 1998, Springer. 5. G. Winkler, K. Hahn, A. Martin, K. Rodenacker, V. Aurich: Noise Reduction in Images: Some Recent Edge-Preserving Methods. Pattern Recognition and Image Analysis, Vol.9, No. 4, 1999,

SPATIO-TEMPORAL DATA ANALYSIS WITH NON-LINEAR FILTERS: BRAIN MAPPING WITH fmri DATA

SPATIO-TEMPORAL DATA ANALYSIS WITH NON-LINEAR FILTERS: BRAIN MAPPING WITH fmri DATA Image Anal Stereol 2000;19:189-194 Original Research Paper SPATIO-TEMPORAL DATA ANALYSIS WITH NON-LINEAR FILTERS: BRAIN MAPPING WITH fmri DATA KARSTEN RODENACKER 1, KLAUS HAHN 1, GERHARD WINKLER 1 AND

More information

2 Michael E. Leventon and Sarah F. F. Gibson a b c d Fig. 1. (a, b) Two MR scans of a person's knee. Both images have high resolution in-plane, but ha

2 Michael E. Leventon and Sarah F. F. Gibson a b c d Fig. 1. (a, b) Two MR scans of a person's knee. Both images have high resolution in-plane, but ha Model Generation from Multiple Volumes using Constrained Elastic SurfaceNets Michael E. Leventon and Sarah F. F. Gibson 1 MIT Artificial Intelligence Laboratory, Cambridge, MA 02139, USA leventon@ai.mit.edu

More information

8/3/2017. Contour Assessment for Quality Assurance and Data Mining. Objective. Outline. Tom Purdie, PhD, MCCPM

8/3/2017. Contour Assessment for Quality Assurance and Data Mining. Objective. Outline. Tom Purdie, PhD, MCCPM Contour Assessment for Quality Assurance and Data Mining Tom Purdie, PhD, MCCPM Objective Understand the state-of-the-art in contour assessment for quality assurance including data mining-based techniques

More information

Segmentation of 3-D medical image data sets with a combination of region based initial segmentation and active surfaces

Segmentation of 3-D medical image data sets with a combination of region based initial segmentation and active surfaces Header for SPIE use Segmentation of 3-D medical image data sets with a combination of region based initial segmentation and active surfaces Regina Pohle, Thomas Behlau, Klaus D. Toennies Otto-von-Guericke

More information

Methodological progress in image registration for ventilation estimation, segmentation propagation and multi-modal fusion

Methodological progress in image registration for ventilation estimation, segmentation propagation and multi-modal fusion Methodological progress in image registration for ventilation estimation, segmentation propagation and multi-modal fusion Mattias P. Heinrich Julia A. Schnabel, Mark Jenkinson, Sir Michael Brady 2 Clinical

More information

Cognition Network Technology for a Fully Automated 3D Segmentation of Liver

Cognition Network Technology for a Fully Automated 3D Segmentation of Liver Cognition Network Technology for a Fully Automated 3D Segmentation of Liver Günter Schmidt, Maria Athelogou, Ralf Schönmeyer, Rene Korn, and Gerd Binnig Definiens AG, Research, Trappentreustr. 1, 80339

More information

Generation of Triangle Meshes from Time-of-Flight Data for Surface Registration

Generation of Triangle Meshes from Time-of-Flight Data for Surface Registration Generation of Triangle Meshes from Time-of-Flight Data for Surface Registration Thomas Kilgus, Thiago R. dos Santos, Alexander Seitel, Kwong Yung, Alfred M. Franz, Anja Groch, Ivo Wolf, Hans-Peter Meinzer,

More information

Interactive segmentation of vascular structures in CT images for liver surgery planning

Interactive segmentation of vascular structures in CT images for liver surgery planning Interactive segmentation of vascular structures in CT images for liver surgery planning L. Wang¹, C. Hansen¹, S.Zidowitz¹, H. K. Hahn¹ ¹ Fraunhofer MEVIS, Institute for Medical Image Computing, Bremen,

More information

Edge-Preserving Denoising for Segmentation in CT-Images

Edge-Preserving Denoising for Segmentation in CT-Images Edge-Preserving Denoising for Segmentation in CT-Images Eva Eibenberger, Anja Borsdorf, Andreas Wimmer, Joachim Hornegger Lehrstuhl für Mustererkennung, Friedrich-Alexander-Universität Erlangen-Nürnberg

More information

Using Probability Maps for Multi organ Automatic Segmentation

Using Probability Maps for Multi organ Automatic Segmentation Using Probability Maps for Multi organ Automatic Segmentation Ranveer Joyseeree 1,2, Óscar Jiménez del Toro1, and Henning Müller 1,3 1 University of Applied Sciences Western Switzerland (HES SO), Sierre,

More information

Segmentation and Grouping

Segmentation and Grouping Segmentation and Grouping How and what do we see? Fundamental Problems ' Focus of attention, or grouping ' What subsets of pixels do we consider as possible objects? ' All connected subsets? ' Representation

More information

Fast 3D Mean Shift Filter for CT Images

Fast 3D Mean Shift Filter for CT Images Fast 3D Mean Shift Filter for CT Images Gustavo Fernández Domínguez, Horst Bischof, and Reinhard Beichel Institute for Computer Graphics and Vision, Graz University of Technology Inffeldgasse 16/2, A-8010,

More information

Prostate Detection Using Principal Component Analysis

Prostate Detection Using Principal Component Analysis Prostate Detection Using Principal Component Analysis Aamir Virani (avirani@stanford.edu) CS 229 Machine Learning Stanford University 16 December 2005 Introduction During the past two decades, computed

More information

Overcompressing JPEG images with Evolution Algorithms

Overcompressing JPEG images with Evolution Algorithms Author manuscript, published in "EvoIASP2007, Valencia : Spain (2007)" Overcompressing JPEG images with Evolution Algorithms Jacques Lévy Véhel 1, Franklin Mendivil 2 and Evelyne Lutton 1 1 Inria, Complex

More information

Performance Evaluation of the TINA Medical Image Segmentation Algorithm on Brainweb Simulated Images

Performance Evaluation of the TINA Medical Image Segmentation Algorithm on Brainweb Simulated Images Tina Memo No. 2008-003 Internal Memo Performance Evaluation of the TINA Medical Image Segmentation Algorithm on Brainweb Simulated Images P. A. Bromiley Last updated 20 / 12 / 2007 Imaging Science and

More information

doi: /

doi: / Yiting Xie ; Anthony P. Reeves; Single 3D cell segmentation from optical CT microscope images. Proc. SPIE 934, Medical Imaging 214: Image Processing, 9343B (March 21, 214); doi:1.1117/12.243852. (214)

More information

Automatic segmentation of the cortical grey and white matter in MRI using a Region Growing approach based on anatomical knowledge

Automatic segmentation of the cortical grey and white matter in MRI using a Region Growing approach based on anatomical knowledge Automatic segmentation of the cortical grey and white matter in MRI using a Region Growing approach based on anatomical knowledge Christian Wasserthal 1, Karin Engel 1, Karsten Rink 1 und André Brechmann

More information

SUPPLEMENTARY FILE S1: 3D AIRWAY TUBE RECONSTRUCTION AND CELL-BASED MECHANICAL MODEL. RELATED TO FIGURE 1, FIGURE 7, AND STAR METHODS.

SUPPLEMENTARY FILE S1: 3D AIRWAY TUBE RECONSTRUCTION AND CELL-BASED MECHANICAL MODEL. RELATED TO FIGURE 1, FIGURE 7, AND STAR METHODS. SUPPLEMENTARY FILE S1: 3D AIRWAY TUBE RECONSTRUCTION AND CELL-BASED MECHANICAL MODEL. RELATED TO FIGURE 1, FIGURE 7, AND STAR METHODS. 1. 3D AIRWAY TUBE RECONSTRUCTION. RELATED TO FIGURE 1 AND STAR METHODS

More information

Artifacts and Textured Region Detection

Artifacts and Textured Region Detection Artifacts and Textured Region Detection 1 Vishal Bangard ECE 738 - Spring 2003 I. INTRODUCTION A lot of transformations, when applied to images, lead to the development of various artifacts in them. In

More information

Computational Medical Imaging Analysis Chapter 4: Image Visualization

Computational Medical Imaging Analysis Chapter 4: Image Visualization Computational Medical Imaging Analysis Chapter 4: Image Visualization Jun Zhang Laboratory for Computational Medical Imaging & Data Analysis Department of Computer Science University of Kentucky Lexington,

More information

STIC AmSud Project. Graph cut based segmentation of cardiac ventricles in MRI: a shape-prior based approach

STIC AmSud Project. Graph cut based segmentation of cardiac ventricles in MRI: a shape-prior based approach STIC AmSud Project Graph cut based segmentation of cardiac ventricles in MRI: a shape-prior based approach Caroline Petitjean A joint work with Damien Grosgeorge, Pr Su Ruan, Pr JN Dacher, MD October 22,

More information

Manifold Learning-based Data Sampling for Model Training

Manifold Learning-based Data Sampling for Model Training Manifold Learning-based Data Sampling for Model Training Shuqing Chen 1, Sabrina Dorn 2, Michael Lell 3, Marc Kachelrieß 2,Andreas Maier 1 1 Pattern Recognition Lab, FAU Erlangen-Nürnberg 2 German Cancer

More information

Modeling and preoperative planning for kidney surgery

Modeling and preoperative planning for kidney surgery Modeling and preoperative planning for kidney surgery Refael Vivanti Computer Aided Surgery and Medical Image Processing Lab Hebrew University of Jerusalem, Israel Advisor: Prof. Leo Joskowicz Clinical

More information

Liver Segmentation in CT Data: A Segmentation Refinement Approach

Liver Segmentation in CT Data: A Segmentation Refinement Approach Liver Segmentation in CT Data: A Segmentation Refinement Approach Reinhard Beichel 12, Christian Bauer 3, Alexander Bornik 3, Erich Sorantin 4, and Horst Bischof 3 1 Dept. of Electrical and Computer Engineering,

More information

Comparison of fiber orientation analysis methods in Avizo

Comparison of fiber orientation analysis methods in Avizo Comparison of fiber orientation analysis methods in Avizo More info about this article: http://www.ndt.net/?id=20865 Abstract Rémi Blanc 1, Peter Westenberger 2 1 FEI, 3 Impasse Rudolf Diesel, Bât A, 33708

More information

Segmenting Lesions in Multiple Sclerosis Patients James Chen, Jason Su

Segmenting Lesions in Multiple Sclerosis Patients James Chen, Jason Su Segmenting Lesions in Multiple Sclerosis Patients James Chen, Jason Su Radiologists and researchers spend countless hours tediously segmenting white matter lesions to diagnose and study brain diseases.

More information

3D Volume Mesh Generation of Human Organs Using Surface Geometries Created from the Visible Human Data Set

3D Volume Mesh Generation of Human Organs Using Surface Geometries Created from the Visible Human Data Set 3D Volume Mesh Generation of Human Organs Using Surface Geometries Created from the Visible Human Data Set John M. Sullivan, Jr., Ziji Wu, and Anand Kulkarni Worcester Polytechnic Institute Worcester,

More information

Fully Automatic Model Creation for Object Localization utilizing the Generalized Hough Transform

Fully Automatic Model Creation for Object Localization utilizing the Generalized Hough Transform Fully Automatic Model Creation for Object Localization utilizing the Generalized Hough Transform Heike Ruppertshofen 1,2,3, Cristian Lorenz 2, Peter Beyerlein 4, Zein Salah 3, Georg Rose 3, Hauke Schramm

More information

Using the Kolmogorov-Smirnov Test for Image Segmentation

Using the Kolmogorov-Smirnov Test for Image Segmentation Using the Kolmogorov-Smirnov Test for Image Segmentation Yong Jae Lee CS395T Computational Statistics Final Project Report May 6th, 2009 I. INTRODUCTION Image segmentation is a fundamental task in computer

More information

Segmentation of Fat and Fascias in Canine Ultrasound Images

Segmentation of Fat and Fascias in Canine Ultrasound Images Segmentation of Fat and Fascias in Canine Ultrasound Images Oleksiy Rybakov 1, Daniel Stromer 1, Irina Mischewski 2 and Andreas Maier 1 1 Pattern Recognition Lab, Friedrich-Alexander-University Erlangen-Nuremberg

More information

GENERAL AUTOMATED FLAW DETECTION SCHEME FOR NDE X-RAY IMAGES

GENERAL AUTOMATED FLAW DETECTION SCHEME FOR NDE X-RAY IMAGES GENERAL AUTOMATED FLAW DETECTION SCHEME FOR NDE X-RAY IMAGES Karl W. Ulmer and John P. Basart Center for Nondestructive Evaluation Department of Electrical and Computer Engineering Iowa State University

More information

Segmentation of Bony Structures with Ligament Attachment Sites

Segmentation of Bony Structures with Ligament Attachment Sites Segmentation of Bony Structures with Ligament Attachment Sites Heiko Seim 1, Hans Lamecker 1, Markus Heller 2, Stefan Zachow 1 1 Visualisierung und Datenanalyse, Zuse-Institut Berlin (ZIB), 14195 Berlin

More information

Texture-Based Detection of Myositis in Ultrasonographies

Texture-Based Detection of Myositis in Ultrasonographies Texture-Based Detection of Myositis in Ultrasonographies Tim König 1, Marko Rak 1, Johannes Steffen 1, Grit Neumann 2, Ludwig von Rohden 2, Klaus D. Tönnies 1 1 Institut für Simulation & Graphik, Otto-von-Guericke-Universität

More information

How and what do we see? Segmentation and Grouping. Fundamental Problems. Polyhedral objects. Reducing the combinatorics of pose estimation

How and what do we see? Segmentation and Grouping. Fundamental Problems. Polyhedral objects. Reducing the combinatorics of pose estimation Segmentation and Grouping Fundamental Problems ' Focus of attention, or grouping ' What subsets of piels do we consider as possible objects? ' All connected subsets? ' Representation ' How do we model

More information

Ensemble registration: Combining groupwise registration and segmentation

Ensemble registration: Combining groupwise registration and segmentation PURWANI, COOTES, TWINING: ENSEMBLE REGISTRATION 1 Ensemble registration: Combining groupwise registration and segmentation Sri Purwani 1,2 sri.purwani@postgrad.manchester.ac.uk Tim Cootes 1 t.cootes@manchester.ac.uk

More information

Preliminaries: Size Measures and Shape Coordinates

Preliminaries: Size Measures and Shape Coordinates 2 Preliminaries: Size Measures and Shape Coordinates 2.1 Configuration Space Definition 2.1 The configuration is the set of landmarks on a particular object. The configuration matrix X is the k m matrix

More information

Automated segmentation methods for liver analysis in oncology applications

Automated segmentation methods for liver analysis in oncology applications University of Szeged Department of Image Processing and Computer Graphics Automated segmentation methods for liver analysis in oncology applications Ph. D. Thesis László Ruskó Thesis Advisor Dr. Antal

More information

The organization of the human cerebral cortex estimated by intrinsic functional connectivity

The organization of the human cerebral cortex estimated by intrinsic functional connectivity 1 The organization of the human cerebral cortex estimated by intrinsic functional connectivity Journal: Journal of Neurophysiology Author: B. T. Thomas Yeo, et al Link: https://www.ncbi.nlm.nih.gov/pubmed/21653723

More information

CHAPTER 3 RETINAL OPTIC DISC SEGMENTATION

CHAPTER 3 RETINAL OPTIC DISC SEGMENTATION 60 CHAPTER 3 RETINAL OPTIC DISC SEGMENTATION 3.1 IMPORTANCE OF OPTIC DISC Ocular fundus images provide information about ophthalmic, retinal and even systemic diseases such as hypertension, diabetes, macular

More information

Rule-Based Ventral Cavity Multi-organ Automatic Segmentation in CT Scans

Rule-Based Ventral Cavity Multi-organ Automatic Segmentation in CT Scans Rule-Based Ventral Cavity Multi-organ Automatic Segmentation in CT Scans Assaf B. Spanier (B) and Leo Joskowicz The Rachel and Selim Benin School of Computer Science and Engineering, The Hebrew University

More information

Interactive Differential Segmentation of the Prostate using Graph-Cuts with a Feature Detector-based Boundary Term

Interactive Differential Segmentation of the Prostate using Graph-Cuts with a Feature Detector-based Boundary Term MOSCHIDIS, GRAHAM: GRAPH-CUTS WITH FEATURE DETECTORS 1 Interactive Differential Segmentation of the Prostate using Graph-Cuts with a Feature Detector-based Boundary Term Emmanouil Moschidis emmanouil.moschidis@postgrad.manchester.ac.uk

More information

3D contour based local manual correction of tumor segmentations in CT scans

3D contour based local manual correction of tumor segmentations in CT scans 3D contour based local manual correction of tumor segmentations in CT scans Frank Heckel a, Jan Hendrik Moltz a, Lars Bornemann a, Volker Dicken a, Hans-Christian Bauknecht b,michaelfabel c, Markus Hittinger

More information

Intraoperative Prostate Tracking with Slice-to-Volume Registration in MR

Intraoperative Prostate Tracking with Slice-to-Volume Registration in MR Intraoperative Prostate Tracking with Slice-to-Volume Registration in MR Sean Gill a, Purang Abolmaesumi a,b, Siddharth Vikal a, Parvin Mousavi a and Gabor Fichtinger a,b,* (a) School of Computing, Queen

More information

Efficient Liver Segmentation exploiting Level-Set Speed Images with 2.5D Shape Propagation

Efficient Liver Segmentation exploiting Level-Set Speed Images with 2.5D Shape Propagation Efficient Liver Segmentation exploiting Level-Set Speed Images with 2.5D Shape Propagation Jeongjin Lee 1, Namkug Kim 2, Ho Lee 1, Joon Beom Seo 2, Hyung Jin Won 2, Yong Moon Shin 2, and Yeong Gil Shin

More information

IMAGE SEGMENTATION. Václav Hlaváč

IMAGE SEGMENTATION. Václav Hlaváč IMAGE SEGMENTATION Václav Hlaváč Czech Technical University in Prague Faculty of Electrical Engineering, Department of Cybernetics Center for Machine Perception http://cmp.felk.cvut.cz/ hlavac, hlavac@fel.cvut.cz

More information

2D Grey-Level Convex Hull Computation: A Discrete 3D Approach

2D Grey-Level Convex Hull Computation: A Discrete 3D Approach 2D Grey-Level Convex Hull Computation: A Discrete 3D Approach Ingela Nyström 1, Gunilla Borgefors 2, and Gabriella Sanniti di Baja 3 1 Centre for Image Analysis, Uppsala University Uppsala, Sweden ingela@cb.uu.se

More information

A Study of Medical Image Analysis System

A Study of Medical Image Analysis System Indian Journal of Science and Technology, Vol 8(25), DOI: 10.17485/ijst/2015/v8i25/80492, October 2015 ISSN (Print) : 0974-6846 ISSN (Online) : 0974-5645 A Study of Medical Image Analysis System Kim Tae-Eun

More information

STEREO BY TWO-LEVEL DYNAMIC PROGRAMMING

STEREO BY TWO-LEVEL DYNAMIC PROGRAMMING STEREO BY TWO-LEVEL DYNAMIC PROGRAMMING Yuichi Ohta Institute of Information Sciences and Electronics University of Tsukuba IBARAKI, 305, JAPAN Takeo Kanade Computer Science Department Carnegie-Mellon

More information

Semantic Context Forests for Learning- Based Knee Cartilage Segmentation in 3D MR Images

Semantic Context Forests for Learning- Based Knee Cartilage Segmentation in 3D MR Images Semantic Context Forests for Learning- Based Knee Cartilage Segmentation in 3D MR Images MICCAI 2013: Workshop on Medical Computer Vision Authors: Quan Wang, Dijia Wu, Le Lu, Meizhu Liu, Kim L. Boyer,

More information

NIH Public Access Author Manuscript Proc IEEE Int Symp Biomed Imaging. Author manuscript; available in PMC 2014 November 15.

NIH Public Access Author Manuscript Proc IEEE Int Symp Biomed Imaging. Author manuscript; available in PMC 2014 November 15. NIH Public Access Author Manuscript Published in final edited form as: Proc IEEE Int Symp Biomed Imaging. 2013 April ; 2013: 748 751. doi:10.1109/isbi.2013.6556583. BRAIN TUMOR SEGMENTATION WITH SYMMETRIC

More information

Norbert Schuff VA Medical Center and UCSF

Norbert Schuff VA Medical Center and UCSF Norbert Schuff Medical Center and UCSF Norbert.schuff@ucsf.edu Medical Imaging Informatics N.Schuff Course # 170.03 Slide 1/67 Objective Learn the principle segmentation techniques Understand the role

More information

10.1 Overview. Section 10.1: Overview. Section 10.2: Procedure for Generating Prisms. Section 10.3: Prism Meshing Options

10.1 Overview. Section 10.1: Overview. Section 10.2: Procedure for Generating Prisms. Section 10.3: Prism Meshing Options Chapter 10. Generating Prisms This chapter describes the automatic and manual procedure for creating prisms in TGrid. It also discusses the solution to some common problems that you may face while creating

More information

Reduction of Metal Artifacts in Computed Tomographies for the Planning and Simulation of Radiation Therapy

Reduction of Metal Artifacts in Computed Tomographies for the Planning and Simulation of Radiation Therapy Reduction of Metal Artifacts in Computed Tomographies for the Planning and Simulation of Radiation Therapy T. Rohlfing a, D. Zerfowski b, J. Beier a, P. Wust a, N. Hosten a, R. Felix a a Department of

More information

A nodal based evolutionary structural optimisation algorithm

A nodal based evolutionary structural optimisation algorithm Computer Aided Optimum Design in Engineering IX 55 A dal based evolutionary structural optimisation algorithm Y.-M. Chen 1, A. J. Keane 2 & C. Hsiao 1 1 ational Space Program Office (SPO), Taiwan 2 Computational

More information

MEDICAL IMAGE NOISE REDUCTION AND REGION CONTRAST ENHANCEMENT USING PARTIAL DIFFERENTIAL EQUATIONS

MEDICAL IMAGE NOISE REDUCTION AND REGION CONTRAST ENHANCEMENT USING PARTIAL DIFFERENTIAL EQUATIONS MEDICAL IMAGE NOISE REDUCTION AND REGION CONTRAST ENHANCEMENT USING PARTIAL DIFFERENTIAL EQUATIONS Miguel Alemán-Flores, Luis Álvarez-León Departamento de Informática y Sistemas, Universidad de Las Palmas

More information

Generation of Hulls Encompassing Neuronal Pathways Based on Tetrahedralization and 3D Alpha Shapes

Generation of Hulls Encompassing Neuronal Pathways Based on Tetrahedralization and 3D Alpha Shapes Generation of Hulls Encompassing Neuronal Pathways Based on Tetrahedralization and 3D Alpha Shapes Dorit Merhof 1,2, Martin Meister 1, Ezgi Bingöl 1, Peter Hastreiter 1,2, Christopher Nimsky 2,3, Günther

More information

Interactive Boundary Detection for Automatic Definition of 2D Opacity Transfer Function

Interactive Boundary Detection for Automatic Definition of 2D Opacity Transfer Function Interactive Boundary Detection for Automatic Definition of 2D Opacity Transfer Function Martin Rauberger, Heinrich Martin Overhoff Medical Engineering Laboratory, University of Applied Sciences Gelsenkirchen,

More information

Semi-automated Basal Ganglia Segmentation Using Large Deformation Diffeomorphic Metric Mapping

Semi-automated Basal Ganglia Segmentation Using Large Deformation Diffeomorphic Metric Mapping Semi-automated Basal Ganglia Segmentation Using Large Deformation Diffeomorphic Metric Mapping Ali Khan 1, Elizabeth Aylward 2, Patrick Barta 3, Michael Miller 4,andM.FaisalBeg 1 1 School of Engineering

More information

Segmentation of Images

Segmentation of Images Segmentation of Images SEGMENTATION If an image has been preprocessed appropriately to remove noise and artifacts, segmentation is often the key step in interpreting the image. Image segmentation is a

More information

EE368 Project: Visual Code Marker Detection

EE368 Project: Visual Code Marker Detection EE368 Project: Visual Code Marker Detection Kahye Song Group Number: 42 Email: kahye@stanford.edu Abstract A visual marker detection algorithm has been implemented and tested with twelve training images.

More information

Edge and local feature detection - 2. Importance of edge detection in computer vision

Edge and local feature detection - 2. Importance of edge detection in computer vision Edge and local feature detection Gradient based edge detection Edge detection by function fitting Second derivative edge detectors Edge linking and the construction of the chain graph Edge and local feature

More information

EE 701 ROBOT VISION. Segmentation

EE 701 ROBOT VISION. Segmentation EE 701 ROBOT VISION Regions and Image Segmentation Histogram-based Segmentation Automatic Thresholding K-means Clustering Spatial Coherence Merging and Splitting Graph Theoretic Segmentation Region Growing

More information

Nonrigid Registration with Adaptive, Content-Based Filtering of the Deformation Field

Nonrigid Registration with Adaptive, Content-Based Filtering of the Deformation Field Nonrigid Registration with Adaptive, Content-Based Filtering of the Deformation Field Marius Staring*, Stefan Klein and Josien P.W. Pluim Image Sciences Institute, University Medical Center Utrecht, P.O.

More information

I How does the formulation (5) serve the purpose of the composite parameterization

I How does the formulation (5) serve the purpose of the composite parameterization Supplemental Material to Identifying Alzheimer s Disease-Related Brain Regions from Multi-Modality Neuroimaging Data using Sparse Composite Linear Discrimination Analysis I How does the formulation (5)

More information

MEDICAL IMAGE COMPUTING (CAP 5937) LECTURE 10: Medical Image Segmentation as an Energy Minimization Problem

MEDICAL IMAGE COMPUTING (CAP 5937) LECTURE 10: Medical Image Segmentation as an Energy Minimization Problem SPRING 07 MEDICAL IMAGE COMPUTING (CAP 97) LECTURE 0: Medical Image Segmentation as an Energy Minimization Problem Dr. Ulas Bagci HEC, Center for Research in Computer Vision (CRCV), University of Central

More information

CS 664 Segmentation. Daniel Huttenlocher

CS 664 Segmentation. Daniel Huttenlocher CS 664 Segmentation Daniel Huttenlocher Grouping Perceptual Organization Structural relationships between tokens Parallelism, symmetry, alignment Similarity of token properties Often strong psychophysical

More information

Storage Efficient NL-Means Burst Denoising for Programmable Cameras

Storage Efficient NL-Means Burst Denoising for Programmable Cameras Storage Efficient NL-Means Burst Denoising for Programmable Cameras Brendan Duncan Stanford University brendand@stanford.edu Miroslav Kukla Stanford University mkukla@stanford.edu Abstract An effective

More information

Fully Automatic Multi-organ Segmentation based on Multi-boost Learning and Statistical Shape Model Search

Fully Automatic Multi-organ Segmentation based on Multi-boost Learning and Statistical Shape Model Search Fully Automatic Multi-organ Segmentation based on Multi-boost Learning and Statistical Shape Model Search Baochun He, Cheng Huang, Fucang Jia Shenzhen Institutes of Advanced Technology, Chinese Academy

More information

DEPTH-ADAPTIVE SUPERVOXELS FOR RGB-D VIDEO SEGMENTATION. Alexander Schick. Fraunhofer IOSB Karlsruhe

DEPTH-ADAPTIVE SUPERVOXELS FOR RGB-D VIDEO SEGMENTATION. Alexander Schick. Fraunhofer IOSB Karlsruhe DEPTH-ADAPTIVE SUPERVOXELS FOR RGB-D VIDEO SEGMENTATION David Weikersdorfer Neuroscientific System Theory Technische Universität München Alexander Schick Fraunhofer IOSB Karlsruhe Daniel Cremers Computer

More information

GeoInterp: Contour Interpolation with Geodesic Snakes Release 1.00

GeoInterp: Contour Interpolation with Geodesic Snakes Release 1.00 GeoInterp: Contour Interpolation with Geodesic Snakes Release 1.00 Rohit R. Saboo, Julian G. Rosenman and Stephen M. Pizer July 1, 2006 University of North Carolina at Chapel Hill Abstract The process

More information

The Anatomical Equivalence Class Formulation and its Application to Shape-based Computational Neuroanatomy

The Anatomical Equivalence Class Formulation and its Application to Shape-based Computational Neuroanatomy The Anatomical Equivalence Class Formulation and its Application to Shape-based Computational Neuroanatomy Sokratis K. Makrogiannis, PhD From post-doctoral research at SBIA lab, Department of Radiology,

More information

A Design Toolbox to Generate Complex Phantoms for the Evaluation of Medical Image Processing Algorithms

A Design Toolbox to Generate Complex Phantoms for the Evaluation of Medical Image Processing Algorithms A Design Toolbox to Generate Complex Phantoms for the Evaluation of Medical Image Processing Algorithms Omar Hamo, Georg Nelles, Gudrun Wagenknecht Central Institute for Electronics, Research Center Juelich,

More information

Automatized & Interactive. Muscle tissues characterization using. Na MRI

Automatized & Interactive. Muscle tissues characterization using. Na MRI Automatized & Interactive Human Skeletal Muscle Segmentation Muscle tissues characterization using 23 Na MRI Noura Azzabou 30 April 2013 What is muscle segmentation? Axial slice of the thigh of a healthy

More information

Invariant Recognition of Hand-Drawn Pictograms Using HMMs with a Rotating Feature Extraction

Invariant Recognition of Hand-Drawn Pictograms Using HMMs with a Rotating Feature Extraction Invariant Recognition of Hand-Drawn Pictograms Using HMMs with a Rotating Feature Extraction Stefan Müller, Gerhard Rigoll, Andreas Kosmala and Denis Mazurenok Department of Computer Science, Faculty of

More information

Chapter 2 Basic Structure of High-Dimensional Spaces

Chapter 2 Basic Structure of High-Dimensional Spaces Chapter 2 Basic Structure of High-Dimensional Spaces Data is naturally represented geometrically by associating each record with a point in the space spanned by the attributes. This idea, although simple,

More information

Introduction to Medical Imaging (5XSA0) Module 5

Introduction to Medical Imaging (5XSA0) Module 5 Introduction to Medical Imaging (5XSA0) Module 5 Segmentation Jungong Han, Dirk Farin, Sveta Zinger ( s.zinger@tue.nl ) 1 Outline Introduction Color Segmentation region-growing region-merging watershed

More information

3D VISUALIZATION OF SEGMENTED CRUCIATE LIGAMENTS 1. INTRODUCTION

3D VISUALIZATION OF SEGMENTED CRUCIATE LIGAMENTS 1. INTRODUCTION JOURNAL OF MEDICAL INFORMATICS & TECHNOLOGIES Vol. 10/006, ISSN 164-6037 Paweł BADURA * cruciate ligament, segmentation, fuzzy connectedness,3d visualization 3D VISUALIZATION OF SEGMENTED CRUCIATE LIGAMENTS

More information

OBJECT SORTING IN MANUFACTURING INDUSTRIES USING IMAGE PROCESSING

OBJECT SORTING IN MANUFACTURING INDUSTRIES USING IMAGE PROCESSING OBJECT SORTING IN MANUFACTURING INDUSTRIES USING IMAGE PROCESSING Manoj Sabnis 1, Vinita Thakur 2, Rujuta Thorat 2, Gayatri Yeole 2, Chirag Tank 2 1 Assistant Professor, 2 Student, Department of Information

More information

Combining PGMs and Discriminative Models for Upper Body Pose Detection

Combining PGMs and Discriminative Models for Upper Body Pose Detection Combining PGMs and Discriminative Models for Upper Body Pose Detection Gedas Bertasius May 30, 2014 1 Introduction In this project, I utilized probabilistic graphical models together with discriminative

More information

Comparison of Vessel Segmentations Using STAPLE

Comparison of Vessel Segmentations Using STAPLE Comparison of Vessel Segmentations Using STAPLE Julien Jomier, Vincent LeDigarcher, and Stephen R. Aylward Computer-Aided Diagnosis and Display Lab, The University of North Carolina at Chapel Hill, Department

More information

Evaluation of Local Filter Approaches for Diffusion Tensor based Fiber Tracking

Evaluation of Local Filter Approaches for Diffusion Tensor based Fiber Tracking Evaluation of Local Filter Approaches for Diffusion Tensor based Fiber Tracking D. Merhof 1, M. Buchfelder 2, C. Nimsky 3 1 Visual Computing, University of Konstanz, Konstanz 2 Department of Neurosurgery,

More information

Automatic Segmentation of Parotids from CT Scans Using Multiple Atlases

Automatic Segmentation of Parotids from CT Scans Using Multiple Atlases Automatic Segmentation of Parotids from CT Scans Using Multiple Atlases Jinzhong Yang, Yongbin Zhang, Lifei Zhang, and Lei Dong Department of Radiation Physics, University of Texas MD Anderson Cancer Center

More information

Automatic Parameter Optimization for De-noising MR Data

Automatic Parameter Optimization for De-noising MR Data Automatic Parameter Optimization for De-noising MR Data Joaquín Castellanos 1, Karl Rohr 2, Thomas Tolxdorff 3, and Gudrun Wagenknecht 1 1 Central Institute for Electronics, Research Center Jülich, Germany

More information

Motion Analysis. Motion analysis. Now we will talk about. Differential Motion Analysis. Motion analysis. Difference Pictures

Motion Analysis. Motion analysis. Now we will talk about. Differential Motion Analysis. Motion analysis. Difference Pictures Now we will talk about Motion Analysis Motion analysis Motion analysis is dealing with three main groups of motionrelated problems: Motion detection Moving object detection and location. Derivation of

More information

Copyright 2007 Society of Photo Optical Instrumentation Engineers. This paper was published in Proceedings of SPIE, volume 6514, Medical Imaging

Copyright 2007 Society of Photo Optical Instrumentation Engineers. This paper was published in Proceedings of SPIE, volume 6514, Medical Imaging Copyright 2007 Society of Photo Optical Instrumentation Engineers. This paper was published in Proceedings of SPIE, volume 6514, Medical Imaging 2007: Computer Aided Diagnosis and is made available as

More information

ADAPTIVE GRAPH CUTS WITH TISSUE PRIORS FOR BRAIN MRI SEGMENTATION

ADAPTIVE GRAPH CUTS WITH TISSUE PRIORS FOR BRAIN MRI SEGMENTATION ADAPTIVE GRAPH CUTS WITH TISSUE PRIORS FOR BRAIN MRI SEGMENTATION Abstract: MIP Project Report Spring 2013 Gaurav Mittal 201232644 This is a detailed report about the course project, which was to implement

More information

Clustering and Visualisation of Data

Clustering and Visualisation of Data Clustering and Visualisation of Data Hiroshi Shimodaira January-March 28 Cluster analysis aims to partition a data set into meaningful or useful groups, based on distances between data points. In some

More information

Optical Crack Detection of Refractory Bricks

Optical Crack Detection of Refractory Bricks ECNDT 2006 - Poster 70 Optical Crack Detection of Refractory Bricks Franz Christian MAYRHOFER, FH OOE Forschungs und Entwicklungs GmbH, Wels, Austria; Kurt NIEL, FH OOE Studienbetriebs GmbH, Wels, Austria

More information

Parameterization of Triangular Meshes with Virtual Boundaries

Parameterization of Triangular Meshes with Virtual Boundaries Parameterization of Triangular Meshes with Virtual Boundaries Yunjin Lee 1;Λ Hyoung Seok Kim 2;y Seungyong Lee 1;z 1 Department of Computer Science and Engineering Pohang University of Science and Technology

More information

Projection and Reconstruction-Based Noise Filtering Methods in Cone Beam CT

Projection and Reconstruction-Based Noise Filtering Methods in Cone Beam CT Projection and Reconstruction-Based Noise Filtering Methods in Cone Beam CT Benedikt Lorch 1, Martin Berger 1,2, Joachim Hornegger 1,2, Andreas Maier 1,2 1 Pattern Recognition Lab, FAU Erlangen-Nürnberg

More information

Atlas Based Segmentation of the prostate in MR images

Atlas Based Segmentation of the prostate in MR images Atlas Based Segmentation of the prostate in MR images Albert Gubern-Merida and Robert Marti Universitat de Girona, Computer Vision and Robotics Group, Girona, Spain {agubern,marly}@eia.udg.edu Abstract.

More information

Computer-Aided Diagnosis in Abdominal and Cardiac Radiology Using Neural Networks

Computer-Aided Diagnosis in Abdominal and Cardiac Radiology Using Neural Networks Computer-Aided Diagnosis in Abdominal and Cardiac Radiology Using Neural Networks Du-Yih Tsai, Masaru Sekiya and Yongbum Lee Department of Radiological Technology, School of Health Sciences, Faculty of

More information

Deviceless respiratory motion correction in PET imaging exploring the potential of novel data driven strategies

Deviceless respiratory motion correction in PET imaging exploring the potential of novel data driven strategies g Deviceless respiratory motion correction in PET imaging exploring the potential of novel data driven strategies Presented by Adam Kesner, Ph.D., DABR Assistant Professor, Division of Radiological Sciences,

More information

MR IMAGE SEGMENTATION

MR IMAGE SEGMENTATION MR IMAGE SEGMENTATION Prepared by : Monil Shah What is Segmentation? Partitioning a region or regions of interest in images such that each region corresponds to one or more anatomic structures Classification

More information

Iterative CT Reconstruction Using Curvelet-Based Regularization

Iterative CT Reconstruction Using Curvelet-Based Regularization Iterative CT Reconstruction Using Curvelet-Based Regularization Haibo Wu 1,2, Andreas Maier 1, Joachim Hornegger 1,2 1 Pattern Recognition Lab (LME), Department of Computer Science, 2 Graduate School in

More information

Chapter 11 Representation & Description

Chapter 11 Representation & Description Chain Codes Chain codes are used to represent a boundary by a connected sequence of straight-line segments of specified length and direction. The direction of each segment is coded by using a numbering

More information

Classification of Subject Motion for Improved Reconstruction of Dynamic Magnetic Resonance Imaging

Classification of Subject Motion for Improved Reconstruction of Dynamic Magnetic Resonance Imaging 1 CS 9 Final Project Classification of Subject Motion for Improved Reconstruction of Dynamic Magnetic Resonance Imaging Feiyu Chen Department of Electrical Engineering ABSTRACT Subject motion is a significant

More information

A MORPHOLOGY-BASED FILTER STRUCTURE FOR EDGE-ENHANCING SMOOTHING

A MORPHOLOGY-BASED FILTER STRUCTURE FOR EDGE-ENHANCING SMOOTHING Proceedings of the 1994 IEEE International Conference on Image Processing (ICIP-94), pp. 530-534. (Austin, Texas, 13-16 November 1994.) A MORPHOLOGY-BASED FILTER STRUCTURE FOR EDGE-ENHANCING SMOOTHING

More information

SuRVoS Workbench. Super-Region Volume Segmentation. Imanol Luengo

SuRVoS Workbench. Super-Region Volume Segmentation. Imanol Luengo SuRVoS Workbench Super-Region Volume Segmentation Imanol Luengo Index - The project - What is SuRVoS - SuRVoS Overview - What can it do - Overview of the internals - Current state & Limitations - Future

More information

Bioimage Informatics

Bioimage Informatics Bioimage Informatics Lecture 13, Spring 2012 Bioimage Data Analysis (IV) Image Segmentation (part 2) Lecture 13 February 29, 2012 1 Outline Review: Steger s line/curve detection algorithm Intensity thresholding

More information