Anesthesia in Cosmetic Surgery

Angiography and Plaque Imaging: Advanced Segmentation Techniques (Biomedical Engineering Series (Boca Raton, Fla.).)......Page 1
Title......Page 2
ANGIOGRAPHY and PLAQUE IMAGING: Advanced Segmentation Techniques......Page 4
Dedication......Page 6
Preface......Page 7
Acknowledgments......Page 12
The Editors......Page 14
Contributors......Page 16
Contents......Page 18
Key Words......Page 0
1.1 Introduction......Page 20
1.2 A Brief Note on MRA Data Acquisition and Prefiltering......Page 32
1.2.1 Time-of-Flight (TOF) MR Angiography......Page 34
1.2.2 Phase Contrast MR Angiography (PC-MRA)......Page 36
1.2.3 Black-Blood Angiography (BBA)......Page 38
1.2.4 Contrast Enhanced MR Angiography (CE-MRA)......Page 39
1.3 Non-Skeleton (Direct) Vascular Segmentation Techniques......Page 41
1.3.1.1 Maximum Intensity Projection and Local Maximum Intensity Projection......Page 42
1.3.1.2 Manual Thresholding for Vascular Segmentation......Page 44
1.3.1.3 Computerized Thresholding and Shading......Page 45
1.3.1.4 EM-Based Automatic Thresholding for Vascular Segmentation......Page 46
1.3.2.1 Moment-Based Approach for Vascular Segmentation......Page 47
1.3.2.2 Vessel Segmentation Using Spline-Based Models......Page 49
1.3.3 Scale-Space Fuzzy Connectedness Technique for Vascular Segmentation......Page 50
1.3.4 Geometric Deformable Models for Vasculature Segmentation......Page 51
1.3.5 Mathematical Morphology Approach for Vessel Segmentation......Page 54
1.3.6 Vasculature Segmentation Using Multi-scale Three-Dimensional Line Filter......Page 55
1.3.7 Skeleton-Based with Multi-scale Vessel Model......Page 58
1.4.1 Skeleton-Based with Scale-Space Edge-Based Fusion......Page 59
1.4.2 Skeleton-Based with Parametric-Based Model Fusion......Page 61
1.4.4 Skeleton-Based Approach Using Differential Geometry......Page 62
1.5 Discussions of Skeleton-Based vs. Non-Skeleton-Based Methods......Page 64
1.5.1 Discussion of the PDE-Based Methods for Vascular Segmentation Algorithms......Page 65
1.5.2 Disadvantages of the Geometric Approaches......Page 67
1.5.3 Discussion of the Scale-Space Issues on Vascular Filtering and Segmentation......Page 69
1.6 Challenges, the Future, and Summary: Vascular Segmentation......Page 70
1.6.2 The Future......Page 71
References......Page 72
2.1 Introduction......Page 88
2.2 Theory and Algorithm of Ellipsoidal Filtering......Page 94
2.2.1 Algorithmic Steps for Ellipsoidal Filtering......Page 96
2.2.2 Experimental Protocol for WBA/BBA Filtering......Page 99
2.3.1 Alexander et al.’s Three-Dimensional Separable Median BBA Filtering Algorithm......Page 101
2.3.2 Results on Alexander et al.’s Technique and Their Pros and Cons......Page 104
2.3.3 Sun et al.’s Local Maximum Mean Filtering Algorithm......Page 109
2.4 MR Angiography Filtering: Implementation Issues......Page 111
2.5.1 Conclusions......Page 112
Acknowledgments......Page 113
References......Page 114
3.1.1 Problem Domain......Page 117
3.2.1 Abdominal......Page 118
3.2.2 Head and Neck......Page 119
3.3.1 Vessel View......Page 120
3.4 Visualization......Page 121
3.4.1 Integrating Visualization and Segmentation......Page 123
3.5.1 Statistical Methods......Page 125
3.5.2.1 Active Contours and Surfaces: Explicit Methods......Page 126
3.5.2.2 Level Set Evolution: Implicit Methods......Page 128
3.5.3 Graph Algorithms......Page 130
3.5.4 Differential Geometry-Based Filtering Approaches......Page 131
3.6 Workflow Constraints......Page 132
3.7.1 Seeded Vessel Segmentation......Page 133
3.7.1.2 Surface Expansion......Page 134
3.7.1.5 The Fermi-Dirac Distribution......Page 136
3.7.1.7 Approximation Methods......Page 137
3.7.2.2 Active Backtracking and System Equilibration......Page 139
3.7.4 Limitations......Page 140
3.7.5 Performance......Page 141
3.8 Cross-Sectional Vessel Boundary Detection by Two-Dimensional Ray Propagation......Page 142
3.8.1 PDE-Based Smoothing Methods......Page 143
3.8.3 Mean-Shift Filtering Along a Vector......Page 145
3.8.4 Mean-Shift-Based Ray Propagation......Page 146
3.8.5 Results......Page 148
3.9 Three-Dimensional Ray Propagation......Page 151
3.9.2 Implementation......Page 152
3.10 Conclusion: From Diagnosis to Screening......Page 153
References......Page 155
4.1 Introduction......Page 161
4.2 Physics of Black-Blood Angiography Image Generation......Page 167
4.2.3 Blood-Flow Dependent Technique: Presaturation......Page 170
4.2.4 Blood-Flow Dependent Technique: Phase Dispersion (Dephasing)......Page 171
4.2.5 Blood-Flow Dependent Technique: Washout Effect......Page 172
4.2.6 Clinically Applied Techniques and Examples: TOF vs. FSE......Page 174
4.2.7 Combined Technique: Double IR Preparation......Page 175
4.2.8 Pros and Cons of the Black-Blood MRA Technique......Page 176
4.3.1 Mask Using Bayesian Techniques: An Application in Black-Blood Orthopedic Images......Page 177
4.3.2 Mask Using Fuzzy Techniques: An Application in Black-Blood Neuro Images......Page 181
4.3.2.1 FCM Algorithm for Mask Generation in Black-Blood Neuro Images......Page 182
4.3.2.2 Recursive Mathematical Morphology with Connected Component Analysis......Page 183
4.4.1 Ellipsoidal Scale-Space Filtering for BBA Vascular Segmentation......Page 184
4.4.2 Algorithmic Steps for BBA Ellipsoidal Filtering......Page 193
4.5 Comparisons: Three-Dimensional Median vs. Three-Dimensional Black-Blood Ellipsoidal Filtering......Page 195
4.6 Implementation Issues in Black-Blood Angiography Filtering......Page 198
4.7.1 Conclusions......Page 203
References......Page 204
5.1 Introduction......Page 211
5.2.1 Automated Isolation of Morphological and Pathological Features in the Fundus......Page 212
5.3 Image Segmentation Using Wavelets......Page 213
5.3.1 Preprocessing......Page 214
5.3.3 Choosing the Right Mother Wavelet......Page 216
5.3.4 Detection of Vessels Using Morlet Wavelets......Page 219
5.3.5 All Together Now: The Segmentation Algorithm......Page 221
5.3.6 Postprocessing Phase......Page 223
5.4 Assessing the Segmentation Results......Page 224
5.4.1 Wavelet Entropy......Page 226
5.4.2 Correlation Dimension......Page 227
5.5 Experimental Results......Page 229
5.6 Discussion......Page 236
Acknowledgments......Page 239
References......Page 240
6.1 Overview......Page 243
6.2 Introduction......Page 244
6.3 Retinal Imaging Background......Page 251
6.4.1 Thresholding......Page 253
6.4.2 Local Gray-Scale Minima-Based Models......Page 255
6.4.3 Combining Local Gray-Scale Minima with Thresholding......Page 256
6.4.4 Hybrid Gray-Scale and Edge-Based Models......Page 257
6.4.5 Combining Hybrid Gray-Scale and Edge-Based Models with Thresholds......Page 258
6.4.6 Models Based on Parallel and Opposite Two-Dimensional Edges......Page 260
6.4.7 Models Based on Gaussian Cross-Sectional Profiles......Page 261
6.4.8 Models for Vessels with a “Hollow” Appearance......Page 263
6.5 Vessel Extraction Approaches......Page 264
6.5.1 Pixel Processing-Based Algorithms......Page 265
6.5.1.1 Morphological Filtering-Based Algorithms......Page 266
6.5.1.2 Methods Based on Edge Detection......Page 269
6.5.2 Exploratory Algorithms......Page 271
6.5.2.2 Fully Automated Tracing Algorithms......Page 272
6.6 RPI-Trace......Page 274
6.7 Algorithm Design Considerations for Real-Time Tracing......Page 279
6.8 Accurate Extraction of Vessel Bifurcations and Crossovers......Page 285
6.8.1 A Model for Vessel Bifurcations and Intersections......Page 288
6.8.1.2 Gathering Information About Neighboring Traces......Page 289
6.8.1.4 Iterative Estimation of Model Parameters......Page 290
6.8.1.5 Simplified Estimation of Landmark Model Parameters......Page 291
6.10 Implementation Methods......Page 293
6.11 Experimental Validation Using Ground Truth Data......Page 295
6.11.1 Ground Truth from Conflicting Observers......Page 296
6.11.2 Multi-observer (Probabilistic) Gold Standard......Page 298
6.11.3 Tracing......Page 300
6.12 Experimental Analysis of Model and Settings for RPI-Trace......Page 302
6.13 Experimental Assessment of the Impact of Landmark Refinement with the ERPR Algorithm......Page 304
6.14 Chapter Summary......Page 306
Acknowledgments......Page 307
References......Page 308
7.2 Background......Page 316
7.2.1 Atherosclerosis......Page 318
7.2.2.2 Ultrasound......Page 319
7.3.1 Characterization Goal......Page 321
7.3.3 Flow, Motion......Page 322
7.4.1 Hardware: Phased-Array Carotid Coil and Headholder......Page 323
7.4.2 Imaging Sequences......Page 325
7.4.2.1 Flow-Enhanced Gradient Echo Sequence......Page 326
7.4.2.2 Flow-Suppressed Spin Echo (or Fast Spin Echo [FSE]) Techniques......Page 327
7.4.2.4 Contrast Enhancement and Quadruple IR Technique......Page 329
7.5.1 Reproducibility......Page 330
7.5.2 Lipid Core/Hemorrhage......Page 331
7.5.3 Fibrous Cap......Page 332
7.5.4 Contrast Enhancement......Page 334
7.5.5 Atherosclerotic Vessel Wall Area Measurement......Page 335
7.6 State-of-the-Art Protocol......Page 337
7.7 Discussion......Page 339
References......Page 340
8.1 Introduction......Page 347
8.2 Quantifying the Vulnerable Plaque......Page 349
8.2.1 Morphological versus Compositional Indexes......Page 350
8.2.2 A Set of Compositional Indexes......Page 351
8.3 Imaging Methods: The Case for MRI......Page 352
8.4 Global Measurements of Composition......Page 355
8.4.1 Vascularity......Page 356
8.4.2 Percent Enhancement......Page 358
8.5 Interactive Analysis of Plaque Composition......Page 360
8.5.1 Contour Tools......Page 361
8.5.2 Labeling......Page 363
8.6 Automated Analysis of Plaque Composition......Page 365
8.6.1 Coil Correction......Page 366
8.6.2 Registration......Page 367
8.6.4 Example......Page 368
8.7 Three-Dimensional Plaque Analysis......Page 370
8.7.1 Surface Model Construction......Page 371
8.8 Conclusion......Page 373
References......Page 376
9.1 Background......Page 381
9.2 Active Contour Model......Page 382
9.2.1 Energy Formulation for Minimal Path Active Contour Model......Page 384
9.2.2 The New Minimal Path Formulation......Page 385
9.2.3 Basic Graph Search Procedure......Page 386
9.2.4 The “Wriggling” Evolution of a Contour......Page 388
9.2.5 Construction of Potential Window......Page 389
9.2.6 Basic Minimal Energy Path Approach......Page 390
9.3 Accurate Lumen Surface Roughness Measurement......Page 392
9.3.1 Contour Representation......Page 393
9.3.2 Graph-Based Triangulation Tiling......Page 394
9.4 Thickness Measurement......Page 397
9.4.2 Tiling Optimization......Page 398
9.4.3 Multi-resolution Tiling......Page 399
9.5 Carotid Artery Morphological Description......Page 400
9.5.1 Formulation of Morphological Descriptors......Page 401
9.5.2.3 Outer-Wall Boundary Descriptors......Page 403
9.5.2.5 Complexity Thick-Lumen Descriptors......Page 404
9.6 Experimental Results......Page 405
Acknowledgment......Page 409
References......Page 410
10.1 Introduction......Page 413
10.2 Vessel Imaging......Page 414
10.3 IVUS Acquisition......Page 416
10.4 Segmentation of IVUS Images......Page 418
10.5 Texture Characterization......Page 428
10.6 Image-Guided Stent Positioning and Follow-Up......Page 431
10.7 Need for Volumetric Measurements and Real Three-Dimensional Reconstruction of the Vessel......Page 436
10.8 Angiogram Analysis......Page 437
10.9 Extraction of Three-Dimensional Measurements from Angiograms......Page 444
10.10 IVUS and Angiogram Registration......Page 450
10.11 IVUS for Abdominal Aorta Aneurysm Analysis......Page 458
10.12 Conclusions......Page 460
References......Page 461
11.1 Introduction......Page 466
11.2 Vascular Image Acquisition Requirements......Page 468
11.2.1 Two-Dimensional Vascular Imaging via Projection......Page 469
11.2.2.1 Three-Dimensional Resolution in Space and Time......Page 472
11.2.2.2 Intensity Inhomogeneities, Contrast Uptake, and Noise......Page 476
11.2.3 Computational Demands......Page 479
11.3.1 Volume Rendering......Page 481
11.3.2 Surface Rendering......Page 484
11.4 Vascular Image Segmentation......Page 485
11.4.1 Dynamic-Scale Centerline and Radius Representations......Page 488
11.4.1.1 Ridge Criteria......Page 489
11.4.1.2 The Importance of Scale......Page 491
11.4.1.3 Modeling a Tube......Page 492
11.4.1.3.1 Step 1: Automatic Vessel Designation......Page 493
11.4.1.3.2 Step 2: Dynamic-Scale Centerline Modeling......Page 494
11.4.1.3.3 Step 3: Radius Estimation......Page 496
11.4.1.3.4 Speed, Accuracy, and Automation......Page 497
11.4.2 Clinical Applications of Vascular Segmentation......Page 499
11.5 Vascular Image Registration......Page 503
11.5.1 Vascular Model-to-Image Matching......Page 505
11.5.2 Clinical Applications of Registration......Page 507
11.6 Conclusion and Future Work......Page 510
References......Page 511
12.1 Future of Skeleton- and Non-Skeleton-Based Segmentation Techniques......Page 515
12.2 The Future of Retinal Fundus Vasculature......Page 516
12.3 The Future of Model-Based Retinal Angiography......Page 519
12.4 Future Directions in MRI of Atherosclerotic Plaque......Page 522
12.4.1 Advances in MRI Technology......Page 523
12.5 The Future of Ultrasound Angiography......Page 524
12.6 Future Directions in Plaque and Aneurysm Assessment......Page 528
12.6.1.2 Assessment of the AAA Rupture Risk......Page 530
12.6.1.4 Endovascular Graft Treatment Planning......Page 531
12.6.2 Conclusions......Page 532
References......Page 533
Key Words in Chapter 2: Scale-Space 3-D Ellipsoidal Filtering for White and Black Blood MRA......Page 536
Key Words in Chapter 3: Three-Dimensional Interactive Vascular Postprocessing Techniques: Industrial Perspective......Page 537
Key Words in Chapter 4: Masking Strategies for Black-Blood Angiography......Page 538
Key Words in Chapter 5: Segmentation of Retinal Fundus Vasculature in Nonmydriatic Camera Images Using Wavelets......Page 539
Key Words in Chapter 6: Automated Model-Based Segmentation, Tracing, and Analysis of Retinal Vasculature from Digital Fundus Images......Page 540
Key Words in Chapter 8: Analysis and Visualization of Atherosclerotic Plaque Composition by MRI......Page 541
KeyWords in Chapter 11: Clinical Applications Involving Vascular Image Segmentation and Registration......Page 542
Key Words in Chapter 12: A Note on Future Research in Vascular and Plaque Segmentation......Page 543