Tavares Computational Vision and Medical Image Processing

Recent Trends
1. Auflage 2010
ISBN: 978-94-007-0011-6
Verlag: Springer-Verlag
Format: PDF
Kopierschutz: Adobe DRM (»Systemvoraussetzungen)

E-Book, Englisch, 349 Seiten

This book contains extended versions of papers presented at the international Conference VIPIMAGE 2009 – ECCOMAS Thematic Conference on Computational Vision and Medical Image, that was held at Faculdade de Engenharia da Universidade do Porto, Portugal, from 14th to 16th of October 2009. This conference was the second ECCOMAS thematic conference on computational vision and medical image processing. It covered topics related to image processing and analysis, medical imaging and computational modelling and simulation, considering their multidisciplinary nature. The book collects the state-of-the-art research, methods and new trends on the subject of computational vision and medical image processing contributing to the development of these knowledge areas.

Autoren/Hrsg.

Tavares, Joao

Inhaltsverzeichnis

1;Preface;6
2;Contents;8
3;Automatic Segmentation of the Optic Radiation Using DTI in Healthy Subjects and Patients with Glaucoma;12
3.1;1 Introduction;13
3.2;2 Interpolation in the Space of Diffusion Tensors;15
3.3;3 Initial Estimation of the Optic Radiation and the Midbrain;16
3.4;4 Segmentation Using a Statistical Level Set Framework;18
3.5;5 Results and Discussion;20
3.6;6 Conclusion and Future Work;23
3.7;References;24
4;Real Time Colour Based Player Tracking in Indoor Sports;27
4.1;1 Introduction;28
4.2;2 Related Work;29
4.3;3 Architecture;30
4.3.1;3.1 Projected Solution;31
4.3.2;3.2 Tested Solution;31
4.4;4 Image Processing;32
4.4.1;4.1 Team Definition;32
4.4.2;4.2 Background Subtraction;33
4.4.3;4.3 Colour Detection;34
4.4.4;4.4 Blob Aggregation and Characterization;34
4.4.5;4.5 Real World Transformation;36
4.4.6;4.6 Player Tracking;37
4.5;5 Results;38
4.5.1;5.1 Overview;38
4.5.2;5.2 Sample Footage;39
4.5.3;5.3 Player Detection;39
4.5.4;5.4 Player Tracking;41
4.6;6 Conclusions and Future Work;44
4.7;References;45
5;Visualization of the Dynamics of the Female Pelvic Floor Reflex and Steady State Function;46
5.1;1 Introduction;47
5.1.1;1.1 Clinical Problem;47
5.1.2;1.2 Anatomical Considerations;47
5.1.3;1.3 Functional Considerations;48
5.1.4;1.4 Contribution of Imaging;48
5.1.5;1.5 Diagnostic Methods;49
5.1.6;1.6 Evaluation of the Dynamic Function of the PF Using 2D Ultrasound Imaging;50
5.2;2 Methods;51
5.2.1;2.1 Coordinate System of the Anatomic Structures;51
5.2.2;2.2 Motion Tracking Algorithms;52
5.2.3;2.3 Image Segmentation Algorithms;53
5.3;3 Results;55
5.3.1;3.1 Quantitative Analysis of the Static Characters of the UVJ-ARA-SP Triangle;57
5.3.2;3.2 Automatic Detection of the UVJ-ARA-SP Triangle;59
5.3.3;3.3 Quantitative Analysis of the Dynamic Characters of the UVJ-ARA-SP Triangle;60
5.3.4;3.4 Quantitative Measurement of Dynamic Parameters of the UVJ-ARA-SP Triangle;61
5.3.5;3.5 The Kinematical Analysis of the Activities of the UVJ-ARA-SP Triangle;63
5.3.6;3.6 Motion Tracking Algorithms;65
5.3.7;3.7 Visualization of the Dynamic Profiles of the Urethra;65
5.3.8;3.8 Visualization of the Timing of the Dynamic Profiles;67
5.4;4 Bio Mechanical Properties of Pelvic Floor Function Using the Vaginal Probe;71
5.4.1;4.1 Temporal/Spatial Visualization;73
5.4.2;4.2 Resting Closure Profiles;73
5.5;5 Discussion;75
5.6;References;80
6;Population Exposure and Impact Assessment: Benefits of Modeling Urban Land Use in Very High Spatial and Thematic Detail;84
6.1;1 Introduction;84
6.2;2 Data and Study Area;85
6.2.1;2.1 Study Area;85
6.2.2;2.2 Remote Sensing Data and Ancillary Space-Related Information;87
6.3;3 Multi-Source Modeling of Functional Urban Patterns;87
6.3.1;3.1 Object Based Image Analysis and Integrated Land Cover Classification;88
6.3.2;3.2 Progressing from Land Cover to Land Use Assessment by Adding Ancillary Space-Related Information;88
6.4;4 Spatial Analysis of Population Distribution Patterns;90
6.5;5 Exposure and Impact Assessment;91
6.5.1;5.1 Population Exposure to Earthquake Hazard;92
6.5.2;5.2 Street Noise Propagation and Affected Population;94
6.6;6 Conclusion and Outlook;95
6.7;References;97
7;Dynamic Radiography Imaging as a Tool in the Design and Validation of a Novel Intelligent Amputee Socket;99
7.1;1 The Need for Novel Socket Designs in a Constantly Increasing Amputee Population;99
7.2;2 Current State-of-the-Art Socket Evaluation Methodologies Are Inefficient in Assessing Trans-Tibial (TT) Socket Problems;100
7.3;3 Integrating Dynamic Radiographic Imaging with Computer-Aided Design and Computational Modeling in Socket Evaluation;103
7.4;4 SMARTsocket: An Example of Integration of Dynamic Imaging, CAD-CAE and FE Methods in Socket Evaluation;105
7.5;5 Conclusion;116
7.6;References;116
8;A Discrete Level Set Approach for Texture Analysis of Microscopic Liver Images;121
8.1;1 Introduction;121
8.2;2 Mathematical Formulation;123
8.2.1;2.1 Discrete Level Set Theory;123
8.2.2;2.2 Texture Analysis of Liver Tissue;125
8.2.3;2.3 The Proposed Algorithm;126
8.3;3 Morphological and Texture Parameters Identification;126
8.4;4 Numerical Results;127
8.5;5 Conclusions;130
8.6;References;131
9;Deformable and Functional Models;132
9.1;1 Introduction;132
9.2;2 Deformable Models;133
9.2.1;2.1 Energy-Minimizing Snakes;133
9.2.2;2.2 Dynamic Snakes;135
9.2.3;2.3 Discretization and Numerical Simulation;136
9.2.4;2.4 Probabilistic (Bayesian) Interpretation;138
9.2.5;2.5 Higher-Dimensional Generalizations;139
9.2.5.1;2.5.1 Deformable Surfaces;139
9.2.6;2.6 Topology-Adaptive Deformable Models;140
9.2.6.1;2.6.1 Topology-Adaptive Snakes;140
9.2.6.2;2.6.2 Topology-Adaptive Deformable Surfaces;142
9.2.7;2.7 Deformable Organisms;143
9.3;3 Functional Models;145
9.3.1;3.1 Facial Simulation;146
9.3.2;3.2 Biomechanically Simulating and Controlling the Neck-Head-Face Complex;146
9.3.3;3.3 Comprehensive Biomechanical Simulation of the Human Body;147
9.4;4 Conclusion;148
9.5;References;149
10;Medical-GiD: From Medical Images to Simulations, 4D MRI Flow Analysis;151
10.1;1 Introduction;151
10.2;2 Methodology;152
10.2.1;2.1 Medical-GiD;152
10.2.2;2.2 Architecture Design;152
10.2.3;2.3 Magnetic Resonance;153
10.2.4;2.4 Segmentation and Meshing for Computational Simulations;154
10.3;3 Results;157
10.3.1;3.1 Aortic Blood Flow Analysis;157
10.3.2;3.2 Blood Flow Velocity Decoding;161
10.3.3;3.3 Numerical Simulation;161
10.4;4 Conclusion;163
10.5;References;165
11;KM and KHM Clustering Techniques for Colour Image Quantisation;167
11.1;1 Introduction;167
11.2;2 Clustering Techniques;168
11.2.1;2.1 KM Technique;168
11.2.2;2.2 KHM Technique;169
11.3;3 Tools for Evaluation of Quantisers;170
11.4;4 KM versus KHM: Comparison Tests;171
11.5;5 Choice of Initialisation Method;177
11.6;6 Empty Clusters;178
11.7;7 Conclusions;179
11.8;References;179
12;Caries Detection in Panoramic Dental X-ray Images;181
12.1;1 Introduction;181
12.1.1;1.1 Dental X-ray;181
12.1.2;1.2 Main Applications;182
12.1.3;1.3 Clinical Environments;182
12.1.4;1.4 Biometrics;182
12.1.5;1.5 Teeth Segmentation;183
12.1.6;1.6 Active Contours Without Edges;184
12.2;2 Main Goal/Motivation;185
12.3;3 Data-Set;186
12.3.1;3.1 Morphological Properties;186
12.4;4 Method;187
12.4.1;4.1 ROI Definition;187
12.4.2;4.2 Jaws Partition;188
12.4.3;4.3 Teeth Gap Valley Detection;189
12.4.4;4.4 Teeth Division;191
12.4.5;4.5 Tooth Segmentation;191
12.4.6;4.6 Dental Caries Classification;192
12.5;5 Results;193
12.5.1;5.1 Segmentation;193
12.5.2;5.2 Classification;193
12.6;6 Conclusion and Further Work;195
12.7;References;195
13;Noisy Medical Image Edge Detection Algorithm Based on a Morphological Gradient Using Uninorms;197
13.1;1 Introduction;197
13.2;2 Fuzzy Morphological Operators and Its Properties;199
13.3;3 The Proposed Edge Detector Algorithm;201
13.4;4 Experimental Results and Analysis;202
13.5;5 Conclusions and Future Work;210
13.6;References;212
14;Leveraging Graphics Hardware for an Automatic Classification of Bone Tissue;214
14.1;1 Introduction;215
14.2;2 Zernike Moments;216
14.2.1;2.1 Optimizations;217
14.2.2;2.2 Input Images;217
14.3;3 Using GPUs to Compute Zernike Moments;218
14.4;4 Performance Analysis;220
14.4.1;4.1 Hardware Resources;220
14.4.2;4.2 GPU Implementation;221
14.4.3;4.3 Performance Against Existing Methods;221
14.5;5 Classification of Bone Tissue Using Zernike Moments;223
14.5.1;5.1 The Biomedical Problem;223
14.5.2;5.2 A Preliminary Selection of Zernike Moments;224
14.5.3;5.3 Searching for the Optimal Vector of Features;225
14.5.3.1;5.3.1 Candidate Vectors;226
14.5.3.2;5.3.2 Classifiers Used;227
14.5.3.3;5.3.3 Training Samples and Input Tiles;227
14.5.3.4;5.3.4 Classification Results;228
14.5.3.5;5.3.5 The Rank of Favorite Zernike Moments;229
14.6;6 Summary and Conclusions;231
14.7;References;232
15;A Novel Template-Based Approach to the Segmentation of the Hippocampal Region;234
15.1;1 Introduction;235
15.2;2 The Pipeline;236
15.2.1;2.1 Images Dataset;236
15.2.2;2.2 Extraction of the Hippocampal Boxes;237
15.2.3;2.3 Selection of Templates;239
15.3;3 Constrained Gaussian Mixture Model Segmentation of the Brain;240
15.4;4 Hippocampal Mask Template Generation;240
15.4.1;4.1 STAPLE;241
15.4.1.1;4.1.1 The Algorithm;242
15.4.2;4.2 Our Strategy;243
15.4.2.1;4.2.1 Initialisation Strategy;243
15.4.2.2;4.2.2 Convergence Check;244
15.4.2.3;4.2.3 Model Parameters;246
15.5;5 Experimental Assessment;246
15.6;6 Conclusions;249
15.7;References;250
16;Model-Based Segmentation and Fusion of 3D Computed Tomography and 3D Ultrasound of the Eye for Radiotherapy Planning;252
16.1;1 Introduction;253
16.2;2 Related Works on Eye Segmentation;254
16.3;3 Eye Segmentation in the Active Contour Framework;256
16.3.1;3.1 Practical Implementation and Optimization Strategy;258
16.3.1.1;3.1.1 Eyeball Segmentation;258
16.3.1.2;3.1.2 Lens Segmentation;258
16.3.1.3;3.1.3 Optimization;259
16.3.2;3.2 Segmentation Results and Validation;260
16.4;4 Image Fusion;262
16.4.1;4.1 Landmark-Based Registration;263
16.4.2;4.2 Object-Based Transformation;265
16.5;5 Discussion;266
16.6;References;267
17;Flow of a Blood Analogue Solution Through Microfabricated Hyperbolic Contractions;269
17.1;1 Introduction;269
17.2;2 Experimental;271
17.2.1;2.1 Microchannel Geometry;271
17.2.2;2.2 Flow Visualization;272
17.2.3;2.3 Rheological Characterization;272
17.3;3 Results;275
17.3.1;3.1 Newtonian Fluid Flow Patterns;275
17.3.2;3.2 Viscoelastic Fluid Flow Patterns;278
17.4;4 Conclusions;281
17.5;References;283
18;Molecular Imaging of Hypoxia Using Genetic Biosensors;284
18.1;1 Introduction;285
18.1.1;1.1 Optical Methods in Molecular Imaging;285
18.1.2;1.2 Bioluminescence Resonant Energy Transfer (BRET);286
18.1.3;1.3 Hypoxia as a Tumoral Aggresivity Marker in Cancer;286
18.2;2 Materials and Methods;287
18.2.1;2.1 Plasmid Construction;287
18.2.2;2.2 Cell Culture;288
18.2.3;2.3 Transfections;288
18.2.4;2.4 Luciferase Assay;288
18.2.5;2.5 Spectrophotometric Profile of the Fusion Protein;289
18.2.6;2.6 Confocal Microscopy;289
18.2.7;2.7 Fluorescence--Bioluminescence In Vivo Assays and Animal Care;289
18.3;3 Results;289
18.3.1;3.1 Outline of the Hypoxia Genetically Encoded Biosensor;289
18.3.2;3.2 Spectrophotometric Characterization of the Fusion Protein;290
18.3.3;3.3 In Vitro Inducible Response of the Biosensor to HIF-1;291
18.3.4;3.4 In Vivo Inducible Response of the Biosensor to HIF-1;292
18.3.5;3.5 In Vivo Bloluminiscence Resonance Energy Transfer Assessment;294
18.4;4 Discussion and Conclusions;295
18.5;References;297
19;Microscale Flow Dynamics of Red Blood Cells in Microchannels: An Experimental and Numerical Analysis;299
19.1;1 Introduction;300
19.2;2 Materials and Methods;300
19.2.1;2.1 Fabrication of the Microchannels;300
19.2.2;2.2 Working Fluids and Geometry of the Bifurcation;302
19.2.3;2.3 Confocal Micro-PTV Experimental Set-Up;303
19.2.4;2.4 Simulation Method;303
19.3;3 Results and Discussion;305
19.4;4 Limitations and Future Directions;309
19.5;5 Conclusions;309
19.6;References;310
20;Two Approaches for Automatic Nuclei Cell Counting in Low Resolution Fluorescence Images;312
20.1;1 Introduction;312
20.2;2 Related Work;314
20.3;3 The Proposed System;315
20.3.1;3.1 Acquisition;316
20.3.2;3.2 Preprocessing;316
20.3.3;3.3 Segmentation;318
20.3.3.1;3.3.1 Thresholding;319
20.3.3.2;3.3.2 Post Processing;319
20.3.4;3.4 Nuclei Counting;319
20.3.4.1;3.4.1 A First Approach, Based on Rules;320
20.3.4.2;3.4.2 The SVM Classifier Approach;321
20.4;4 Implementation and Tests;325
20.4.1;4.1 Implementation;325
20.4.2;4.2 Test Protocol;325
20.4.3;4.3 Results;326
20.5;5 Conclusion;326
20.6;References;326
21;Cerebral Aneurysms: A Patient-Specific and Image-Based Management Pipeline;328
21.1;1 Introduction;329
21.2;2 Patient-Specific and Image-Based Data Processing Pipeline;330
21.3;3 Anatomical Modeling: From Medical Images to Anatomical Models;331
21.4;4 Morphological Analysis: From Anatomical Models to Shape Descriptors;333
21.5;5 Morphodynamic Analysis: Evaluating Temporal Changes in Medical Images;333
21.5.1;5.1 Quantification of Volumetric Changes;334
21.5.2;5.2 Wall Motion Estimation;335
21.6;6 Structural Analysis: Estimating Aneurysm Wall Mechanical Properties;337
21.7;7 Computational Hemodynamic Analysis: From Anatomical Models to Personalized Flow Descriptors;338
21.8;8 Endovascular Device Modeling: From Anatomical Models to Treatment Assessment;339
21.8.1;8.1 Virtual Stenting;340
21.8.2;8.2 Virtual Coiling;341
21.9;9 Discussion;343
21.10;References;346