دانلود کتاب Handbook of Speckle Filtering and Tracking in Cardiovascular Ultrasound Imaging and Video (Healthcare Technologies)

فهرست مطالب :

Cover\nContents\nPreface\nGuide to book contents\nPart I. Introduction to speckle noise in ultrasound imaging and video\n 1 A brief review of ultrasound imaging and video\n 1.1 Revisiting the notion of ultrasound\n 1.2 Basics of ultrasound imaging: the pulse-echo sensing principle\n 1.2.1 A-mode sensing (amplitude mode)\n 1.2.2 M-mode mapping (motion mode)\n 1.2.3 B-mode imaging (brightness mode)\n 1.3 Imaging considerations: from frame rate to spatial resolution\n 1.3.1 Minimum pulse repetition interval\n 1.3.2 Frame rate\n 1.3.3 Depth (or axial) resolution\n 1.3.4 Lateral resolution\n 1.3.5 Penetration depth\n 1.4 Ultrasound imaging hardware\n 1.4.1 Array transducers\n 1.4.2 System electronics\n 1.4.3 Scanner appearance\n 1.4.4 Portable systems\n 1.5 The Doppler mode\n 1.5.1 Doppler spectrogram\n 1.5.2 Color flow imaging\n 1.5.3 Power Doppler imaging\n 1.6 New trend toward high-frame-rate imaging\n 1.6.1 Current progress in high-frame-rate imaging research\n 1.6.2 Concurrent developments in new imaging hardware\n 1.6.3 Recent advances in high-frame-rate cardiovascular ultrasound\n 1.7 Conclusion\n References\n 2 Speckle physics\n 2.1 Introduction\n 2.2 Speckle observations\n 2.2.1 Speckle in optics\n 2.2.2 Mapping optical speckle concepts to ultrasonic imaging\n 2.3 Speckle as a 2D random walk\n 2.3.1 First-order statistics\n 2.3.2 Sums of speckle patterns\n 2.3.3 Speckle pattern plus a nonrandom phasor\n 2.3.4 Nonuniform phase distribution\n 2.4 Speckle in ultrasonic imaging\n 2.4.1 Random ultrasound scattering in 1D\n 2.4.2 Second-order speckle statistics in ultrasound\n 2.4.3 Partially developed speckle and speckle from few scatterers\n 2.5 Effect of postprocessing on first-and second-order statistics\n 2.6 Summary\n Acknowledgments\n References\n 3 Statistical models for speckle noise and Bayesian deconvolution of ultrasound images\n 3.1 Statistical analysis of speckle noise\n 3.1.1 Statistical models for radio-frequency signals\n 3.1.1.1 Gaussian distribution\n 3.1.1.2 KRF distribution\n 3.1.1.3 Generalized Gaussian distribution\n 3.1.1.4 α-Stable distributions\n 3.1.2 Statistical models for envelope signals\n 3.1.2.1 Rayleigh distribution\n 3.1.2.2 Rice distribution\n 3.1.2.3 K distribution\n 3.1.2.4 Homodyned-K distribution\n 3.1.2.5 Nakagami distribution\n 3.1.2.6 α-Rayleigh distribution\n 3.1.3 Statistical models for B-mode image\n 3.1.4 Brief review of statistical despeckling techniques\n 3.1.4.1 Image filtering\n 3.1.4.2 Compounding\n 3.2 Bayesian method for US image deconvolution\n 3.2.1 Bayesian model for joint deconvolution and segmentation\n 3.2.1.1 Likelihood\n 3.2.1.2 Prior distributions\n 3.2.1.3 Joint posterior distribution\n 3.2.2 Sampling the posterior and computing Bayesian estimators\n 3.2.2.1 Hybrid Gibbs sampler\n 3.2.2.2 Parameter estimation\n 3.2.3 Experimental results\n 3.3 Conclusions\n References\n 4 Summary\nPart II. Speckle filtering\n 5\rIntroduction to speckle filtering\n 5.1 The problem of filtering in medical imaging\n 5.2 Important issues about speckle filtering\n 5.2.1 On the ultimate goal of speckle filtering\n 5.2.1.1 Complete speckle elimination\n 5.2.1.2 Selective speckle elimination\n 5.2.2 On the necessity of an accurate speckle model\n 5.2.3 Practical implementation, filter parameters and noise estimation\n 5.2.3.1 Some practical implementation issues\n 5.2.3.2 Estimation of the coefficient of variation\n 5.2.4 Evaluation and validation of speckle filtering\n 5.2.5 On the similarity between SAR and ultrasound images\n 5.3 Some final remarks\n Acknowledgments\n References\n 6 An overview of despeckle-filtering techniques\n 6.1 An overview of despeckle-filtering techniques\n 6.2 Selected despeckle-filtering applications in ultrasound imaging and video\n References\n 7 Linear despeckle filtering\n 7.1 First-order statistics filtering\n 7.2 Local statistics filtering with higher moments\n 7.3 Homogeneous mask area filtering\n 7.4 Despeckle filtering evaluation on an artificial carotid artery image\n 7.5 Despeckle filtering evaluation on a phantom image\n 7.6 Despeckle filtering evaluation on real ultrasound images and video\n 7.7 Summary findings on despeckle filtering evaluation\n References\n 8 Nonlinear despeckle filtering\n 8.1 Filtering based on local windows\n 8.1.1 Median filter\n 8.1.2 Gamma filter\n 8.1.3 Region-oriented schemes\n 8.2 Nonlocal means schemes\n 8.3 Speckle filtering based on partial differential equations\n 8.3.1 Diffusion filters\n 8.3.1.1 Original formulation\n 8.3.1.2 Speckle-adapted diffusion filtering\n 8.3.2 Total-variation methods\n 8.4 Homomorphic filtering\n 8.5 Bilateral filters\n 8.6 Geometric filtering\n 8.7 Other filtering methodologies\n 8.8 Some final remarks\n Acknowledgments\n References\n 9 Wavelet despeckle filtering\n 9.1 Introduction\n 9.2 Discrete wavelet transform\n 9.3 Limitations of DWT and its improvements in de-noising\n 9.4 Dual tree-complex wavelet transform\n 9.5 DT-CWT and shift-invariance\n 9.6 CWT and directional selectivity\n 9.7 Filter implementation of DT-CWT\n 9.8 Practical algorithm\n 9.9 Results and discussions\n 9.10 Conclusions\n References\n 10 A comparative evaluation on linear and nonlinear despeckle filtering techniques\n 10.1 Despeckle filtering evaluation of carotid plaque imaging based on texture analysis\n 10.1.1 Distance measures\n 10.1.2 Univariate statistical analysis\n 10.1.3 The kNN classifier\n 10.1.4 Image and video quality and visual evaluation\n 10.2 Despeckle filtering based on texture analysis (discussion)\n 10.3 Image despeckle filtering based on visual quality evaluation (discussion)\n 10.4 Despeckle filtering evaluation on carotid plaque video based on texture analysis\n 10.5 Video despeckle filtering based on texture analysis and visual quality evaluation (discussion)\n 10.6 Concluding remarks and future directions\n References\n 11 Summary and future directions\n 11.1 Summary on despeckle filtering\n 11.2 Future directions\n References\nPart III. Speckle tracking\n 12 Introduction to speckle tracking in ultrasound video\n 12.1 M-mode\n 12.1.1 Methods\n 12.1.2 Applications\n 12.2 Doppler imaging\n 12.2.1 Method\n 12.2.2 Limitations\n 12.3 Tissue Doppler imaging\n 12.3.1 Method\n 12.3.2 Applications\n 12.3.3 TDI-based strain (rate) imaging\n 12.3.4 Limitations\n 12.4 Ultrasound elastography\n 12.4.1 Method\n 12.4.2 Speckle tracking\n 12.4.3 RF-based block matching\n 12.4.4 Lateral displacements\n 12.4.5 Developments\n 12.4.6 Applications in breast\n 12.4.7 Other applications\n References\n 13 Principles of speckle tracking\n 13.1 General principles\n 13.2 Classification of speckle tracking techniques\n 13.2.1 Input data type\n 13.2.2 Data dimensionality\n 13.2.3 Temporal tracking strategy\n 13.3 Overview of speckle tracking techniques\n 13.3.1 Doppler-based methods and 1D motion estimators\n 13.3.1.1 Time-shift (or time-delay) estimators\n 13.3.1.2 Phase-shift estimators\n 13.3.2 Optical flow methods\n 13.3.3 Registration-based methods\n 13.3.4 Biomechanical models\n 13.3.5 Statistical models\n 13.3.6 Segmentation-based methods\n 13.4 Determinants of speckle tracking performance\n 13.4.1 Spatial resolution\n 13.4.1.1 Point-spread function\n 13.4.1.2 Sampling criteria\n 13.4.1.3 High frequency imaging\n 13.4.1.4 Transverse oscillations beamforming\n 13.4.1.5 Directional beamforming\n 13.4.1.6 Other beamforming strategies\n 13.4.2 Temporal resolution\n 13.4.2.1 Optimal frame rate\n 13.4.2.2 Intrinsic frame rate trade-offs\n 13.4.2.3 Fast imaging sequences\n 13.4.3 Other factors\n 13.4.3.1 Out-of-plane motion\n 13.4.3.2 Image quality\n 13.4.3.3 Cramer–Rao lower bound\n 13.4.3.4 Tissue type\n 13.4.3.5 Algorithm parameter tuning\n References\n 14 Techniques for speckle tracking: block matching\n 14.1 Introduction\n 14.1.1 Strain imaging: an overview\n 14.1.2 Terminology\n 14.2 1-D speckle tracking and strain imaging\n 14.2.1 Data types\n 14.2.2 Similarity measures\n 14.2.3 Sub-sample displacement estimation\n 14.2.4 Window size\n 14.2.5 De-correlation\n 14.2.6 Re-correlation\n 14.2.6.1 Regularization of cross-correlation functions\n 14.2.6.2 Iterative approaches\n 14.2.6.3 Sub-sample alignment\n 14.2.6.4 Strain-based stretching\n 14.3 Multi-dimensional displacement estimation\n 14.3.1 From line to block matching\n 14.3.2 Multi-dimensional cross-correlation\n 14.3.3 2-D window sizes\n 14.3.3.1 Re-correlation approaches\n 14.4 Resolution\n 14.5 Regularization\n 14.6 Strain estimation\n 14.6.1 Strain measures\n 14.6.1.1 Strain vs. strain rate: the need for tracking\n 14.6.2 Strain vs. local strain\n 14.7 In vivo challenges\n 14.7.1 Mismatch between US propagation direction and tissue strain\n 14.7.2 Anistropy and non-linearity\n 14.8 Elastography\n References\n 15 Techniques for tracking: image registration\n 15.1 Ultrasound image registration: speckle tracking\n 15.2 Similarity model\n 15.2.1 Feature-based image registration\n 15.2.2 Intensity-based image registration\n 15.2.3 Maximum likelihood approach\n 15.3 Transformation model\n 15.3.1 Rigid transformation\n 15.3.2 Nonrigid transformation according to the physical model\n 15.3.2.1 Elastic transformations\n 15.3.2.2 Transformation models based on flow theory\n 15.3.2.3 Fluid flow transformations\n 15.3.2.4 Optical flow\n 15.3.2.5 Diffusion\n 15.3.3 Parametric nonrigid transformation\n 15.3.3.1 Radial basis functions\n 15.3.3.2 B-splines\n 15.3.4 Regularization\n 15.3.5 Diffeomorphic and inverse transformation\n 15.3.5.1 Diffeomorphism by using a variational approach\n 15.4 Optimization strategy\n 15.5 Influence of speckle tracking strategies for motion and strain estimation\n Acknowledgments\n References\n 16 Cardiac strain estimation\n List of acronyms\n 16.1 Myocardial strain imaging: rationale\n 16.2 Myocardial strain: definitions\n 16.2.1 Myocardial strain\n 16.3 Cardiac strain estimation in practice\n 16.4 The effect of smoothness on strain analysis\n 16.5 Factors affecting strain estimation\n 16.5.1 Image quality\n 16.5.2 Modality\n 16.5.3 Vendor software and software version\n 16.5.4 Methodology of estimation\n 16.5.5 Acquisition parameters\n 16.6 How should strain be estimated?\n 16.6.1 Physiological aspects and concerns\n 16.6.2 Processing aspects and concerns\n 16.6.3 Signal to noise aspects and concerns\n 16.6.4 Recommendations\n 16.7 Tracking quality and reliability index\n 16.8 Clinical application of cardiac strain echocardiography\n 16.9 Summary\n References\n 17 Combined techniques of filtering and speckle tracking\n Abstract\n 17.1 Introduction\n 17.2 De-speckling\n 17.3 Feature tracking\n 17.4 Parametric images\n 17.5 Deconvolution\n 17.6 Beamforming\n 17.7 Conclusion\n Acknowledgements\n References\n 18 Summary and future directions\n 18.1 Automation\n References\nPart IV. Selected applications\n 19 Segmentation of the carotid artery IMT in ultrasound\n 19.1 Introduction\n 19.2 Ultrasound and carotid artery characteristics\n 19.2.1 Properties of ultrasound images\n 19.2.2 Carotid artery structure and appearance in ultrasound images\n 19.3 Carotid artery recognition\n 19.3.1 Techniques for automatic carotid artery localization\n 19.3.1.1 Shape priors\n 19.3.1.2 Texture and classification\n 19.3.1.3 Pixel intensity and/or local statistics\n 19.4 Carotid wall and final IMT segmentation\n 19.4.1 Edge-tracking and gradient-based techniques\n 19.4.2 Dynamic programming techniques\n 19.4.3 Active contours (snakes)\n 19.4.4 Transform-based and modeling approaches\n 19.4.5 Data mining techniques\n 19.5 Performance measurement and comparison\n 19.6 Discussion and final remarks\n References\n 20 Ultrasound carotid plaque video segmentation\n 20.1 Introduction\n 20.2 Methodology and materials used\n 20.2.1 Acquisition of ultrasound videos and manual delineation of atherosclerotic plaque\n 20.2.2 Video normalization and speckle reduction filtering of ultrasound videos\n 20.2.3 Plaque contour initialization and snakes segmentation\n 20.2.4 M-mode image generation, boundary extraction, state identification and manual delineation\n 20.2.5 Evaluation of the segmentation method and state diagram\n 20.3 Results\n 20.4 Discussion\n 20.4.1 Limitations of the video segmentation method\n 20.5 Concluding remarks\n References\n 21 Ultrasound asymptomatic carotid plaque image analysis for the prediction of the risk of stroke\n 21.1 Introduction\n 21.2 Collected data\n 21.3 Imaging feature sets\n 21.3.1 Statistical features\n 21.3.2 Spatial gray level dependence matrices\n 21.3.3 Morphological analysis\n 21.4 Risk modeling\n 21.4.1 Classifiers\n 21.4.2 Evaluation\n 21.5 Results\n 21.6 Conclusions\n References\n 22 3D segmentation and texture analysis of the carotid arteries\n 22.1 Introduction\n 22.2 3D carotid ultrasound imaging\n 22.2.1 Advantages of 3D US\n 22.2.2 Mechanical systems for 3D imaging of the carotid arteries\n 22.2.3 Free-hand 3D US imaging\n 22.3 Quantitative analysis of 3D carotid US images\n 22.3.1 Texture analysis\n 22.3.1.1 Texture feature calculation\n 22.3.1.2 Feature selection\n 22.3.1.3 Classification\n 22.3.1.4 Results\n 22.3.2 Semiautomated segmentation algorithms of 3D US carotid images\n 22.3.3 2D and 3D methods for segmenting LIB from 3D US images\n 22.3.4 2D methods that segment both the LIB and MAB from 3D US images\n 22.3.4.1 MAB segmentation\n 22.3.4.2 LIB segmentation\n 22.3.4.3 3D methods that segment both LIB and MAB from 3D US images\n 22.3.5 Segmentation algorithms of carotid plaque from 3D US images\n 22.3.5.1 Manual segmentation of plaque from 3D US images\n 22.3.5.2 Semiautomated segmentation of plaque from 3D US images\n 22.4 Local quantification of carotid atherosclerosis based on 3D US images\n 22.4.1 3D vessel-wall-plus-plaque thickness (VWT) map\n 22.4.2 2D Carotid template\n 22.4.2.1 Arc-length scaling (AL) approach\n 22.5 Optimization of correspondence by minimizing the description length\n 22.5.1 Role of DL minimization to improve reproducibility of 3D US VWT measurements\n 22.5.2 Novel biomarker based on 2D carotid template\n 22.6 Future perspectives\n References\n 23 Carotid artery mechanics assessed by ultrasound\n 23.1 Introduction\n 23.1.1 Anatomy\n 23.1.2 Tissue composition\n 23.1.3 Atherosclerosis\n 23.1.4 Treatment of stenotic arteries\n 23.2 Mechanical behavior of carotid arteries\n 23.2.1 Loading of arteries\n 23.2.2 Deformation of arteries\n 23.2.3 Pressure–diameter relation\n 23.2.3.1 Compliance and distensibility\n 23.2.4 Stress–strain behavior\n 23.2.4.1 Elasticity moduli and stiffness\n 23.2.4.2 Pulse wave velocity\n 23.2.4.3 Stiffness vs. elasticity\n 23.2.5 Nonlinearity\n 23.2.6 Anisotropy and viscoelasticity\n 23.3 US-based assessment of carotid mechanics\n 23.3.1 Motion estimation\n 23.3.1.1 Compliance and distensibility estimation\n 23.3.2 Elastometry\n 23.3.3 Pulse wave velocity imaging\n 23.3.4 Shear wave elastography\n 23.3.5 Strain imaging\n 23.3.6 (Inverse) finite element modeling\n References\n 24 Carotid artery wall motion and strain analysis using tracking\n Abstract\n 24.1 Introduction\n 24.2 Methods for motion and strain analysis\n 24.3 Estimation of motion and strain of the carotid artery in health and disease\n 24.4 Discussion and future perspectives\n References\n 25 IVUS tracking: advantages and disadvantages of intravascular ultrasound in the detection of artery geometrical features and plaque type morphology\n Abstract\n 25.1 Introduction\n 25.2 Background\n 25.2.1 Physical principles of IVUS imaging\n 25.2.2 IVUS acquisition systems\n 25.2.3 IVUS disadvantages (artifacts-problems)\n 25.2.4 IVUS artifacts\n 25.2.4.1 Non-Uniform Rotational Distortion (NURD) and motion artifacts\n 25.2.4.2 Ring-down artifacts\n 25.2.4.3 Guide wire artifacts\n 25.2.4.4 Blood speckles\n 25.2.4.5 Obliquity, eccentricity and vessel curvature problems\n 25.2.4.6 Spatial orientation problem\n 25.3 Speckle noise in IVUS images\n 25.3.1 Model for speckle noise\n 25.3.2 Need for denoising\n 25.3.3 Wavelet transform denoising methods\n 25.3.4 Curvelet transform denoising methods\n 25.4 Segmentation techniques for border identification\n 25.4.1 Edge-tracking and gradient-based techniques\n 25.4.2 Active contour-based techniques\n 25.4.3 Statistical-and probabilistic-based techniques\n 25.4.4 Multiscale expansion-based techniques\n 25.5 Plaque characterization\n 25.5.1 Methodologies developed for plaque characterization using gray-scale IVUS\n 25.5.2 Methodologies developed for plaque characterization using the backscatter IVUS signal\n 25.6 IVUS-based hybrid imaging\n 25.6.1 Fusion of IVUS and coronary angiography\n 25.6.2 Fusion of IVUS and coronary computed tomography\n 25.7 Limitations of IVUS usability\n 25.8 Future perspectives of IVUS\n 25.8.1 Future trends in 3D IVUS reconstruction\n 25.8.2 Future technical development\n 25.9 Conclusions\n References\n 26 Introduction to speckle tracking in cardiac ultrasound imaging\n 26.1 Speckle formation and speckle tracking\n 26.2 Basic principles of speckle tracking\n 26.3 Speckle-tracking echocardiography\n 26.4 Clinical utility of global longitudinal strain in speckle-tracking echocardiography\n 26.5 Echocardiographic particle image velocimetry (Echo-PIV)\n 26.6 Potential clinical utility of vortex flow imaging in echo-PIV\n 26.7 Color Doppler as an alternative or complementary to speckle tracking\n 26.8 Potential benefits of high-frame-rate echocardiography\n 26.9 Toward volumetric speckle-tracking echocardiography\n 26.10 Historical and clinical conclusion\n References\n 27 Assessment of systolic and diastolic heart failure\n Abstract\n 27.1 Clinical phenotypes\n 27.2 Applying basic principles of myocardial deformation\n 27.3 Accuracy, reproducibility, and normal values\n 27.4 Pathophysiology of heart failure\n 27.5 Early diagnosis\n 27.6 Aetiology\n 27.7 Prognosis of heart failure\n 27.8 Left atrial and right ventricular function\n 27.9 Conclusions – imaging in heart failure\n References\n 28 Myocardial elastography and electromechanical wave imaging\n Abstract\n 28.1 Myocardial elastography\n 28.1.1 Introduction\n 28.1.2 Mechanical deformation of normal and ischemic or infarcted myocardium\n 28.1.3 Myocardial elastography\n 28.1.3.1 2D strain estimation and imaging\n 28.1.3.2 3D strain estimation and imaging\n 28.1.3.3 PBME performance assessment\n 28.1.3.4 Compounding\n 28.1.4 Simulations\n 28.1.5 Phantoms\n 28.1.6 Myocardial ischemia and infarction detection in canines in vivo\n 28.1.6.1 Ischemic model\n 28.1.6.2 Infarct model\n 28.1.7 Validation of myocardial elastography against CT angiography\n 28.2 Electromechanical wave imaging\n 28.2.1 Cardiac arrhythmias\n 28.2.2 Clinical diagnosis of atrial arrhythmias\n 28.2.3 Treatment of atrial arrhythmias\n 28.2.4 Electromechanical wave imaging\n 28.2.4.1 The cardiac electromechanics\n 28.2.4.2 Electromechanical wave imaging\n 28.2.4.3 Treatment guidance capability of EWI\n 28.2.5 Imaging the electromechanics of the heart\n 28.2.6 EWI sequences\n 28.2.6.1 The ACT sequence\n 28.2.6.2 The TUAS sequence\n 28.2.6.3 Single-heartbeat EWI and optimal strain estimation\n 28.2.7 Characterization of atrial arrhythmias in canines in vivo\n 28.2.7.1 Electrical mapping\n 28.2.7.2 Validation\n 28.2.8 EWI in normal human subjects and with arrhythmias\n Acknowledgments\n References\n 29 Summary and future directions\n 29.1 Summary on selected applications\n 29.2 Future directions\n References\nIndex\nBack Cover