Handbook of Medical Image Computing and Computer Assisted Intervention (The MICCAI Society book Series)

Handbook of Medical Image Computing and Computer Assisted Intervention presents important advanced methods and state-of-the art research in medical image computing and computer assisted intervention, providing a comprehensive reference on current technical approaches and solutions, while also offering proven algorithms for a variety of essential medical imaging applications. This book is written primarily for university researchers, graduate students and professional practitioners (assuming an elementary level of linear algebra, probability and statistics, and signal processing) working on medical image computing and computer assisted intervention. Presents the key research challenges in medical image computing and computer-assisted intervention Written by leading authorities of the Medical Image Computing and Computer Assisted Intervention (MICCAI) Society Contains state-of-the-art technical approaches to key challenges Demonstrates proven algorithms for a whole range of essential medical imaging applications Includes source codes for use in a plug-and-play manner Embraces future directions in the fields of medical image computing and computer-assisted intervention Cover......Page 1 HANDBOOK OFMEDICAL IMAGECOMPUTING ANDCOMPUTERASSISTEDINTERVENTION......Page 4 Copyrigh......Page 5 Contents......Page 6 Contributors......Page 17 Acknowledgment......Page 27 1.1 Introduction......Page 28 1.2.1 Physics-based image synthesis......Page 29 1.2.2 Classification-based synthesis......Page 30 1.2.3 Registration-based synthesis......Page 32 1.2.4 Example-based synthesis......Page 33 1.2.5 Scan normalization in MRI......Page 36 1.3.1 Superresolution reconstruction......Page 39 1.3.2 Single-image deconvolution......Page 41 1.3.3 Example-based superresolution......Page 42 1.4 Conclusion......Page 45 References......Page 46 2 Machine learning for image reconstruction......Page 52 2.1 Inverse problems in imaging......Page 53 2.2 Unsupervised learning in image reconstruction......Page 57 2.3 Supervised learning in image reconstruction......Page 59 Nonconvex regularization......Page 61 Bi-level optimization......Page 62 Convolutional neural networks as regularization......Page 63 2.3.2 Learning an iterative reconstruction model......Page 64 Example: Single-coil MRI reconstruction Schlemper2018......Page 65 2.3.3 Deep learning for image and data enhancement......Page 71 2.3.4 Learning a direct mapping......Page 73 2.3.5 Example: Comparison between learned iterative reconstruction and learned postprocessing......Page 74 2.4 Training data......Page 75 2.5 Loss functions and evaluation of image quality......Page 76 2.6 Discussion......Page 80 References......Page 81 3.1 Introduction......Page 92 3.1.1 Prior work: segmentation vs. detection......Page 93 3.1.2 FCN for pixel-to-pixel transformations......Page 94 3.2.1 Lesion candidate detection via a fully convolutional network architecture......Page 95 3.2.1.1 FCN candidate generation results......Page 97 3.2.2 Superpixel sparse-based classification for false-positives reduction......Page 98 3.2.3.1 Data......Page 100 3.2.3.2 Comparative system performance......Page 101 3.3 Fully convolutional network for CT to PET synthesis to augment malignant liver lesion detection......Page 104 3.3.1 Related work......Page 105 3.3.2.1 Training data preparation......Page 106 3.3.2.2 The networks......Page 107 3.3.2.3 SUV-adapted loss function......Page 108 3.3.3.2 Experimental setting......Page 109 3.3.3.3 Liver lesion detection using the virtual-PET......Page 110 3.4 Discussion and conclusions......Page 111 References......Page 114 4.1 Overview......Page 118 4.3 Lung CAD systems......Page 119 4.4.1 Lung nodule......Page 120 Hessian-based approach......Page 121 4.4.2 Ground Glass Opacity (GGO) nodule......Page 122 4.5 Diffuse lung disease......Page 123 4.5.1 Emphysema......Page 124 4.6.1 Airway......Page 125 4.6.2 Blood vessel segmentation in the lung......Page 127 4.6.4 Lung lobe segmentation......Page 129 References......Page 131 5.1 Introduction......Page 135 5.2.1 Text mining......Page 136 5.2.2 Disease classification......Page 137 5.3.1.1 Architecture......Page 140 5.3.1.1.3 Negation and uncertainty detection......Page 141 5.3.1.2 Evaluation of NegBio......Page 142 5.3.3.1.1 Unified DCNN framework......Page 143 5.3.3.1.2 Weakly-supervised pathology localization......Page 145 5.3.3.2 Evaluation......Page 146 5.4 Case study 2: text mining in pathology reports and images......Page 148 5.4.2 Language model......Page 149 5.4.3 Dual-attention model......Page 150 5.4.4 Image prediction......Page 151 5.4.5 Evaluation......Page 152 5.5 Conclusion and future work......Page 155 References......Page 156 6 Multiatlas segmentation......Page 162 6.1 Introduction......Page 163 6.2 History of atlas-based segmentation......Page 164 6.2.1 Atlas generation......Page 165 6.2.3 Registration......Page 166 6.2.4 Atlas selection......Page 167 6.2.5.2 Rater modeling......Page 168 6.2.5.3 Bayesian / generative models......Page 169 6.2.6.3 Markov Random Field (MRF)......Page 170 6.3.1 Problem definition......Page 171 6.3.2 Voting label fusion......Page 172 6.3.3 Statistical label fusion......Page 173 6.3.5 Spatial STAPLE......Page 175 6.3.7 Nonlocal spatial STAPLE......Page 176 6.3.8 E-step: estimation of the voxel-wise label probability......Page 177 6.3.9 M-step: estimation of the performance level parameters......Page 178 6.5 Multiatlas segmentation using machine learning......Page 179 6.8 Challenges and applications......Page 180 6.9 Unsolved problems......Page 181 Glossary......Page 182 References......Page 183 7.1.1 Generative adversarial network......Page 190 7.1.2 Deep image-to-image network......Page 192 7.2 Segmentation using an adversarial image-to-image network......Page 194 7.2.1 Experiments......Page 196 7.3 Volumetric domain adaptation with intrinsic semantic cycle consistency......Page 197 7.3.1 Methodology......Page 200 7.3.1.2 Volumetric domain adaptation with cycle consistency......Page 201 7.3.2 Experiments......Page 203 7.3.3 Conclusions......Page 205 References......Page 206 8.1 Introduction......Page 208 8.2.1 Medical image synthesis......Page 211 8.2.2 Image segmentation......Page 212 8.3.2 Generative adversarial network......Page 213 8.3.4 Problems in unpaired volume-to-volume translation......Page 214 8.4.1 Volume-to-volume cycle consistency......Page 215 8.4.3 Multimodal volume segmentation......Page 216 8.5.1 Architecture......Page 217 8.6.1 Dataset......Page 219 8.6.2 Cross-domain translation evaluation......Page 220 8.6.3 Segmentation evaluation......Page 221 8.6.4 Gap between synthetic and real data......Page 224 8.7 Conclusions......Page 225 References......Page 226 9.1 Introduction......Page 230 9.2.1.1 Point-based representation......Page 232 9.2.1.3 Identity map representation......Page 233 9.2.1.5 Heat map representation......Page 234 9.2.1.6 Discrete action map representation......Page 235 9.2.2 Action classification for landmark detection......Page 236 9.2.2.1 Method......Page 237 9.2.2.2 Dataset & experimental setup......Page 238 9.2.2.3 Qualitative and quantitative results......Page 240 9.3.1 Shape representation......Page 242 Global context posterior......Page 244 MMSE estimate for landmark location......Page 245 Shape initialization using robust model alignment......Page 246 9.3.2.3 Comparison with other methods......Page 247 9.3.2.4 Experimental results......Page 248 References......Page 252 10.1 Introduction......Page 255 10.2.1 Electron microscopy image segmentation......Page 257 10.2.2 Nuclei segmentation......Page 258 10.3.1 Deep multilevel contextual network......Page 259 10.3.2 Regularization with auxiliary supervision......Page 260 10.4.1.1 2012 ISBI EM segmentation......Page 261 10.4.3.1 Qualitative evaluation......Page 262 10.4.3.2 Quantitative evaluation metrics......Page 263 10.4.3.3 Results comparison without postprocessing......Page 264 10.4.3.4 Results comparison with postprocessing......Page 265 10.4.4.2 Quantitative evaluation metrics......Page 266 10.4.4.3 Quantitative results and comparison......Page 267 Acknowledgment......Page 268 References......Page 269 11.1 Introduction......Page 272 11.2.1 Initial mesh......Page 274 11.2.3 Cost function design......Page 275 11.2.4 Geometric constraints and priors......Page 277 11.2.5 Graph optimization......Page 278 11.3 Just-enough interaction......Page 279 11.4 Retinal OCT segmentation......Page 280 11.5 Coronary OCT segmentation......Page 283 11.6 Knee MR segmentation......Page 288 11.7 Modular application design......Page 291 11.8 Conclusion......Page 292 References......Page 293 12.1 Introduction......Page 296 12.1.1 Deformable models for cardiac modeling......Page 297 12.1.2 Learning based cardiac segmentation......Page 298 Network architecture......Page 300 Modified deep layer aggregation network......Page 301 Dataset and evaluation metrics......Page 303 Results......Page 304 12.3 Shape refinement by sparse shape composition......Page 306 12.4 3D modeling......Page 308 References......Page 311 13.1 Challenges of motion discontinuities in medical imaging......Page 316 13.2 Sliding preserving regularization for Demons......Page 319 Demons with bilateral filtering......Page 320 GIFTed Demons......Page 322 13.2.2.1 Graph-based regularization for demons......Page 324 13.3 Discrete optimization for displacements......Page 326 13.3.1 Energy terms for discrete registration......Page 328 13.3.2 Practical concerns and implementation details for 3D discrete registration......Page 329 13.3.3 Parameterization of nodes and displacements......Page 330 13.3.3.1 Efficient inference of regularization......Page 331 13.4 Image registration for cancer applications......Page 334 13.5 Conclusions......Page 336 References......Page 337 14.1 Introduction......Page 342 14.2.1 Learning initialized deformation field......Page 345 14.2.2 Learning intermediate image......Page 347 14.2.3 Learning image appearance......Page 348 14.3 Machine-learning-based multimodal registration......Page 349 14.3.1 Learning similarity metric......Page 351 14.3.2 Learning common feature representation......Page 353 14.3.3 Learning appearance mapping......Page 354 14.4.1 Learning similarity metric......Page 355 14.4.2 Learning preliminary transformation parameters......Page 357 14.4.3 End-to-end learning for deformable registration......Page 358 References......Page 361 15 Imaging biomarkers in Alzheimer's disease......Page 366 15.1 Introduction......Page 367 15.2.1.1 Grey matter assessment......Page 368 15.2.1.2 White matter damage......Page 369 15.2.2.1 Functional imaging......Page 370 15.2.2.2 Molecular imaging......Page 371 15.3.2 Biomarkers extraction: from visual scales to automated processes......Page 372 15.3.3 Automated biomarker extraction: behind the scene......Page 374 15.3.4 Automated methodological development validation......Page 375 15.4.1 Practical use......Page 376 15.4.2 Biomarkers' path to validation......Page 377 15.4.3 Current challenges......Page 378 15.5.1.1 Spatial patterns of abnormality - from global to local......Page 379 15.5.2 Longitudinal vs cross-sectional......Page 381 15.5.2.1 Challenges in longitudinal analyses......Page 382 15.6.1.2 Standardization initiatives, challenges and open-source data......Page 383 15.6.2.2 Ever-increasing potential of AI technologies: reproduction, combination, discovery......Page 384 15.6.3 Longitudinal prediction, simulation and ethical considerations......Page 385 References......Page 386 16.1 Introduction......Page 402 16.2.1 The ENIGMA project......Page 405 16.2.3 Harmonization of multisite neuroimaging data......Page 406 16.3 Unsupervised pattern learning for dimensionality reduction of neuroimaging data......Page 408 16.3.1 Finding imaging patterns of covariation......Page 409 16.4 Supervised classification based imaging biomarkers for disease diagnosis......Page 410 16.4.2 Classification of schizophrenia patients in multisite large cohorts......Page 411 16.5.1 Brain development index......Page 412 16.5.2 Imaging patterns of brain aging......Page 413 16.6 Deep learning in neuroimaging analysis......Page 415 16.7 Revealing heterogeneity of imaging patterns of brain diseases......Page 416 16.8 Conclusions......Page 417 References......Page 418 17.1 Introduction......Page 423 17.2 Cardiac imaging......Page 424 17.3 Cardiac shape and function......Page 426 17.3.1 Left ventricular mass......Page 427 17.3.3 Remodeling......Page 429 17.4.1 Wall motion analysis......Page 431 17.4.2 Myocardial strain......Page 432 17.4.3 Dyssynchrony......Page 433 17.5.1 Coronary artery disease......Page 434 17.5.2 Myocardial perfusion......Page 435 17.5.3 Blood flow......Page 436 17.6.1 Tissue characterization......Page 438 17.6.2 Fiber architecture......Page 440 17.7 Population-based cardiac image biomarkers......Page 441 References......Page 442 18 Radiomics......Page 451 18.1 Introduction......Page 452 18.2.1 Introduction......Page 453 18.2.3 Imaging data collection......Page 454 18.2.5 Conclusion......Page 455 18.3.2 Segmentation methods......Page 456 18.3.4 Conclusion......Page 458 18.4.2.1 Morphological features......Page 459 Filter based......Page 460 Gray level matrix features......Page 461 18.4.4 Feature extraction......Page 462 18.5.1 Introduction......Page 463 18.5.3 Machine learning......Page 464 18.5.4 Deep learning......Page 465 18.6.2 Training, validation and evaluation set......Page 466 18.6.3.1 Cross-validation......Page 467 18.6.4 Evaluation metrics......Page 468 18.7.1 Introduction......Page 469 18.7.2 Data storage and sharing......Page 470 18.7.3 Feature toolboxes......Page 471 18.7.5 Pipeline standardization......Page 472 18.8 Conclusion......Page 473 References......Page 474 19.1 A different way to use context......Page 479 19.2 Feature selection and ensembling......Page 481 19.3.1 Inference......Page 482 19.3.2 Training......Page 484 Optimization......Page 485 Leaf predictions......Page 486 Effect of model parameters......Page 487 19.3.3 Integrating context......Page 488 19.4 Applications......Page 489 19.4.1 Detection and localization......Page 490 19.4.2 Segmentation......Page 491 19.4.3 Image-based prediction......Page 493 19.4.4 Image synthesis......Page 495 19.4.5 Feature interpretation......Page 496 19.4.6 Algorithmic variations......Page 497 References......Page 498 20.1 Introduction......Page 503 20.2 Neural networks......Page 504 20.2.1 Loss function......Page 505 20.2.2 Backpropagation......Page 506 20.3 Convolutional neural networks......Page 509 20.3.1 Convolutions......Page 510 Equivariance......Page 511 20.3.2 Nonlinearities......Page 512 20.3.3 Pooling layers......Page 513 20.4 CNN architectures for classification......Page 514 20.5.1 Data standardization and augmentation......Page 517 20.5.2 Optimizers and learning rate......Page 518 20.5.3 Weight initialization and pretrained networks......Page 519 20.5.4 Regularization......Page 520 20.6 Future challenges......Page 521 References......Page 522 21.1 From feedforward to recurrent......Page 524 21.1.1 Simple motivating example......Page 525 21.1.3 Simple RNNs......Page 526 21.1.4 Representation power of simple RNNs......Page 527 21.2 Modeling with RNNs......Page 528 21.2.1 Discriminative sequence models......Page 529 21.2.3 RNN-based encoder-decoder models......Page 530 21.3.1 The chain rule for ordered derivatives......Page 531 21.3.2 The vanishing gradient problem......Page 532 21.3.4 Teacher forcing......Page 534 21.4 Long short-term memory and gated recurrent units......Page 535 21.5 Example applications of RNNs at MICCAI......Page 538 References......Page 539 22 Deep multiple instance learning for digital histopathology......Page 541 22.1 Multiple instance learning......Page 542 22.2 Deep multiple instance learning......Page 544 22.3 Methodology......Page 545 22.4.1 Instance-based approach......Page 546 22.4.2 Embedding-based approach......Page 547 22.5 MIL pooling functions......Page 548 22.5.3 LSE......Page 550 22.5.5 Attention mechanism......Page 551 22.5.7 Flexibility......Page 552 22.6 Application to histopathology......Page 553 22.6.1.1 Cropping......Page 554 Color decomposition......Page 555 Color normalization......Page 556 22.6.2 Performance metrics......Page 557 22.6.2.2 Precision, recall and F1-score......Page 558 22.6.2.3 Receiver Operating Characteristic Area Under Curve......Page 559 22.6.3.1 Experimental setup......Page 560 22.6.3.2 Colon cancer......Page 562 22.6.3.3 Breast cancer......Page 563 References......Page 565 23.1 Introduction......Page 567 23.2.1 Objective functions......Page 569 23.2.2 The latent space......Page 572 23.2.4 GAN architectures......Page 573 23.3 Adversarial methods for image domain translation......Page 574 23.3.1 Training with paired images......Page 575 23.3.2 Training without paired images......Page 576 23.4 Domain adaptation via adversarial training......Page 578 23.5 Applications in biomedical image analysis......Page 579 23.5.1 Sample generation......Page 580 23.5.2 Image synthesis......Page 581 23.5.3 Image quality enhancement......Page 582 23.5.4 Image segmentation......Page 584 23.5.5 Domain adaptation......Page 585 23.5.6 Semisupervised learning......Page 587 23.6 Discussion and conclusion......Page 588 References......Page 590 24 Linear statistical shape models and landmark location......Page 595 24.2.1 Representing structures with points......Page 596 24.2.2 Comparing two shapes......Page 597 24.2.4 Aligning a set of shapes......Page 598 24.2.5 Building linear shape models......Page 599 24.2.5.1 Choosing the number of modes......Page 600 24.2.5.3 Matching a model to known points......Page 601 24.2.6 Analyzing shapes......Page 602 24.2.8 Limitations of linear models......Page 603 24.2.10.1 Level set representations......Page 604 24.2.11 3D models......Page 605 24.3.1 Exhaustive methods: searching for individual points......Page 606 24.3.1.3 Discriminative approaches......Page 607 24.3.1.4 Regression-based approaches......Page 608 24.3.2 Alternating approaches......Page 610 24.3.2.1 Constrained local models......Page 611 24.3.3 Iterative update approaches......Page 612 24.3.3.1 Updating parameters......Page 613 24.3.3.2 Regression-based updates......Page 614 24.A.1 Computing modes when fewer samples than ordinates......Page 615 24.A.3 Closest point on an ellipsoid......Page 616 References......Page 617 25.1 Introduction: a three-way partnership between humans, technology, and information to improve patient care......Page 619 25.2 The information flow in computer-integrated interventional medicine......Page 623 25.2.2 Patient-specific models......Page 624 25.2.4 Treatment planning......Page 625 25.2.5 Intervention......Page 626 25.2.7 Multipatient information and statistical analysis......Page 627 25.3 Intraoperative systems for CIIM......Page 628 25.3.1 Intraoperative imaging systems......Page 629 25.3.2 Navigational trackers......Page 631 25.3.3 Robotic devices......Page 632 25.3.4 Human-machine interfaces......Page 634 25.4 Emerging research themes......Page 635 References......Page 636 26 Technology and applications in interventional imaging: 2D X-ray radiography/fluoroscopy and 3D cone-beam CT......Page 645 26.1.1 Production of X-rays for fluoroscopy and CBCT......Page 646 26.1.2 Large-area X-ray detectors for fluoroscopy and cone-beam CT......Page 648 26.1.4.1 Detector corrections / image preprocessing......Page 651 26.1.4.2 Postprocessing......Page 652 26.1.5.1 Measurement of fluoroscopic dose......Page 653 26.1.5.2 Reference dose levels......Page 654 26.2.1.1 Geometrical calibration......Page 655 26.2.1.2 I0 calibration......Page 657 26.2.1.3 Other correction factors......Page 658 26.2.2.1 Filtered backprojection......Page 660 26.2.2.2 Emerging methods: optimization-based (iterative) image reconstruction (OBIR)......Page 664 26.2.2.3 Emerging methods: machine learning methods for cone-beam CT......Page 666 26.2.3.1 Measurement of dose in CBCT......Page 667 26.2.3.2 Reference dose levels......Page 668 26.3.1 Mobile systems: C-arms, U-arms, and O-arms......Page 669 26.3.2 Fixed-room C-arm systems......Page 671 26.4 Applications......Page 674 26.4.1.1 Neurological interventions......Page 675 26.4.1.2 Body interventions (oncology and embolization)......Page 677 26.4.2 Interventional cardiology......Page 679 26.4.3 Surgery......Page 681 References......Page 684 27 Interventional imaging: MR......Page 692 27.1 Motivation......Page 693 27.2.1 Design, operation, and safety of an interventional MRI suite......Page 694 27.2.2.1 Needles and biopsy guns......Page 696 27.2.3 Visualization requirements......Page 697 27.2.4 Intraprocedural guidance......Page 698 27.2.4.1 Passive tracking......Page 699 27.2.4.2 Active tracking - radiofrequency coils......Page 700 27.2.4.5 Optical tracking......Page 701 27.2.6 MR elastography......Page 702 27.3.1 Applications in oncology......Page 703 27.3.1.2 Clinical workflow......Page 704 27.3.1.3 MR-guided biopsies......Page 705 27.3.1.4 MR-guided thermal ablations......Page 706 27.3.2.2 Intraoperative MRI and laser interstitial thermal therapy......Page 708 27.3.2.3 Safety considerations......Page 709 References......Page 711 28 Interventional imaging: Ultrasound......Page 719 28.2 Ultrasound-guided cardiac interventions......Page 720 28.2.1 Cardiac ultrasound imaging technology......Page 721 28.2.2.1 Reconstructed 3D imaging......Page 722 28.3 Ultrasound data manipulation and image fusion for cardiac applications......Page 723 28.3.2 Integration of ultrasound imaging with surgical tracking......Page 724 28.3.3 Fusion of ultrasound imaging via volume rendering......Page 725 28.4 Ultrasound imaging in orthopedics......Page 726 28.4.1.1 Segmentation methods using image intensity and phase information......Page 727 28.4.1.3 Incorporation of bone shadow region information to improve segmentation......Page 728 28.5.1 Fluoroscopy & TEE-guided aortic valve implantation......Page 729 28.5.3 Model-enhanced US-guided intracardiac interventions......Page 730 28.5.4 ICE-guided ablation therapy......Page 731 28.5.5 Image-guided spine interventions......Page 732 Acknowledgments......Page 733 References......Page 734 29 Interventional imaging: Vision......Page 739 29.1.1.1 Endoscope types......Page 740 29.1.1.2 Advances in endoscopic imaging......Page 741 29.1.2 Microscopy......Page 742 29.2.1 Calibration and preprocessing......Page 743 29.2.1.1 Preprocessing......Page 744 29.2.2.1 Stereo reconstruction......Page 745 29.2.2.2 Simultaneous Localization and Mapping......Page 746 29.2.2.3 Shape-from-X......Page 747 29.2.3 Registration......Page 748 29.2.3.2 Surface-based registration......Page 749 29.3.1 Detection......Page 750 29.3.1.2 Phase detection......Page 751 29.4.2 Tissue characterization......Page 752 29.5 Discussion......Page 754 References......Page 756 30 Interventional imaging: Biophotonics......Page 764 30.1 A brief introduction to light-tissue interactions and white light imaging......Page 765 30.2 Summary of chapter structure......Page 766 30.3 Fluorescence imaging......Page 767 30.4 Multispectral imaging......Page 770 30.5 Microscopy techniques......Page 773 30.6 Optical coherence tomography......Page 777 30.7 Photoacoustic methods......Page 778 30.8 Optical perfusion imaging......Page 781 30.9 Macroscopic scanning of optical systems and visualization......Page 782 References......Page 784 31.1 Introduction......Page 793 31.2 Target registration error estimation for paired measurements......Page 794 31.3.1 Electromagnetic tracking system......Page 796 31.3.2 Optical tracking system......Page 797 31.4 Stylus calibration......Page 798 31.5 Template-based calibration......Page 800 31.6 Ultrasound probe calibration......Page 801 31.7 Camera hand-eye calibration......Page 803 31.8 Conclusion and resources......Page 806 References......Page 807 32.1 Background and motivation......Page 811 32.2 General concepts......Page 812 32.3.1 Background......Page 814 32.3.3 Planning workflow......Page 815 32.3.4 Planning system......Page 816 32.3.5 Evaluation and validation......Page 819 32.4.1 Background......Page 820 32.4.2 Placement constraints......Page 822 32.4.3 Constraint solving......Page 823 32.4.4 Evaluation and validation......Page 824 32.5 Future challenges......Page 827 References......Page 829 33.1 HCI for medical imaging vs clinical interventions......Page 832 33.1.1 HCI for diagnostic queries (using medical imaging)......Page 833 33.1.2 HCI for planning, guiding, and executing imperative actions (computer-assisted interventions)......Page 834 33.2 Human-computer interfaces: design and evaluation......Page 835 33.3 What is an interface?......Page 836 33.5 Position inputs (free-space pointing and navigation interactions)......Page 837 33.6 Direct manipulation vs proxy-based interactions (cursors)......Page 838 33.8 Selection (object-based interactions)......Page 839 33.9 Quantification (object-based position setting)......Page 840 33.10 User interactions: selection vs position, object-based vs free-space......Page 841 33.12 Language-based control (text commands or spoken language)......Page 843 33.13 Image-based and workspace-based interactions: movement and selection events......Page 845 33.14 Task representations for image-based and intervention-based interfaces......Page 848 33.15 Design and evaluation guidelines for human-computer interfaces: human inputs are computer outputs - the system design must respect perceptual capacities and constraints......Page 849 33.16 Objective evaluation of performance on a task mediated by an interface......Page 851 References......Page 853 34.1 Introduction......Page 855 34.2 Precision positioning......Page 856 34.3 Master-slave system......Page 858 34.4 Image guided robotic tool guide......Page 860 34.5 Interactive manipulation......Page 862 34.6 Articulated access......Page 864 34.7 Untethered microrobots......Page 866 34.8 Soft robotics......Page 867 References......Page 869 35 System integration......Page 875 35.2 System design......Page 876 35.2.2 Design approaches......Page 877 35.3.1 Middleware......Page 878 35.3.1.2 Data serialization......Page 879 35.3.1.3 Robot Operating System (ROS)......Page 880 35.3.2.1 Requirements......Page 881 35.3.2.2 Overview of existing application frameworks......Page 883 35.4.1 Software configuration management......Page 885 35.4.3 Documentation......Page 887 35.4.4 Testing......Page 888 35.5.1.1 DVRK system architecture......Page 889 35.5.1.2 dVRK I/O layer......Page 891 35.5.1.3 DVRK real-time control layer......Page 892 35.5.1.4 DVRK ROS interface......Page 893 35.5.1.6 DVRK with augmented reality HMD......Page 895 35.5.2 SlicerIGT based interventional and training systems......Page 896 35.5.2.1 3D Slicer module design......Page 897 35.5.2.2 Surgical navigation system for breast cancer resection......Page 898 35.5.2.3 Virtual/augmented reality applications......Page 900 35.6 Conclusions......Page 902 References......Page 903 36.1 Introduction......Page 906 36.2 Definitions......Page 907 36.3 Useful researcher characteristics for clinical translation......Page 909 36.3.1 Comfort zone......Page 910 36.3.3 Embracing change......Page 911 36.3.5 Selection of a clinical translatable idea......Page 912 36.3.7 Regulatory approval......Page 913 36.4.2 Clinical research partners and generation of the hypothesis......Page 914 36.4.3 Development of basic tools......Page 915 36.4.5 Clinical research......Page 916 36.4.7 Actions based on lessons learned......Page 917 36.5 Conclusions......Page 918 References......Page 919 37.1 Introduction......Page 921 37.2 Assessment......Page 923 37.2.1 Rating by expert reviewers......Page 924 37.2.2 Real-time spatial tracking......Page 925 37.2.3 Automatic video analysis......Page 927 37.2.4 Crowdsourcing......Page 928 37.3.1 Feedback in complex procedures......Page 929 37.3.2 Learning curves and performance benchmarks......Page 930 37.4 Simulated environments......Page 931 37.4.2 Synthetic models......Page 932 37.4.3 Box trainers......Page 933 37.4.4 Virtual reality......Page 935 37.5 Shared resources......Page 936 References......Page 937 38.1 Concept of surgical data science (SDS)......Page 942 38.2 Clinical context for SDS and its applications......Page 945 Automating intelligent surgical assistance......Page 946 Improving measurement of surgical outcomes......Page 947 Integrating data science into the surgical care pathway......Page 948 Data sources......Page 949 Creating labeled data and dealing with sparsely annotated data:......Page 950 Ontologies and semantic models......Page 951 Inference and machine learning......Page 952 Pervasive data capture......Page 953 Models of surgeon performance......Page 954 Finding good use cases......Page 955 References......Page 956 39.1 Introduction......Page 964 39.2.1 Geometry extraction from medical images: segmentation......Page 966 39.2.2 Finite element mesh generation......Page 967 39.2.3 Image as a computational biomechanics model: meshless discretization......Page 969 39.3 Biomechanics informs image analysis: computational biomechanics model as image registration tool......Page 972 39.3.1 Biomechanics-based image registration: problem formulation......Page 974 39.3.2.1 Neuroimage registration......Page 975 39.3.2.2 Magnetic resonance (MR) image registration for intracranial electrode localization for epilepsy treatment......Page 976 39.4 Discussion......Page 980 References......Page 983 40.1 Introduction to computer assisted interventions......Page 989 40.1.3 Application domain for interventions......Page 990 Relevance......Page 991 Flexibility......Page 992 Usability......Page 993 40.2.1 Robotics......Page 994 40.2.2 Augmented reality and advanced visualization/interaction concepts......Page 998 40.2.3 Artificial intelligence - data-driven decision support......Page 1000 40.3 Translational challenge......Page 1001 Certification / regulatory affairs......Page 1002 Financing......Page 1003 40.4 Simulation......Page 1004 Simulation within the healthcare innovation pathway......Page 1005 Simulation-based assessment......Page 1006 Prototyping......Page 1007 Training......Page 1009 Replacing old knowledge with new knowledge......Page 1012 Intraoperative training and assistance......Page 1013 References......Page 1014 Index......Page 1023 Back Cover......Page 1054