Speckle Tracking Performance in High Frame-Rate Ultrasound Imaging

高速超音波成像的發展有助於提昇斑點追蹤技術的效能。這兩種技術的結合(簡稱為高速成像斑點追蹤技術)能夠突破傳統量測速度的都普勒方法之角度與最大量測速度限制；此外，由於高速成像能夠縮小斑點追蹤演算法的搜尋範圍，所以高速成像斑點追蹤亦能夠節省運算量。因此，本研究的目的即為檢驗上述的假說並評估高速成像斑點追蹤技術的效能。要影響斑點失真與高速成像斑點追蹤技術效能的因素有：影像訊雜比、一致的速度分佈、不一致的速度分佈。因此，我們做了兩種實驗與一種模擬(包含整體性的運動與流體運動)來評估上述因素對高速成像斑點追蹤技術效能的影響。我們採用運算量較低的絕對差和斑點追蹤演算法；高速成像的部分是用平面波成像來實行以避免成像過程中可能造成的掃描效應。速成像斑點追蹤技術的效能評估結果顯示：在影像訊雜比夠高的情況下，此技術能達到高準確度的速度估測結果；此外在提高成像率的同時，斑點追蹤技術的運算量也大大降低。因此，有效率、準確度高的高速成像斑點追蹤技術是可行的。High frame-rate imaging has the potential to improve performance of speckle tracking. Compared with Doppler techniques, with which the maximum detectable velocity is limited and it is flow angle dependent, speckle tracking does not suffer from these limitations. With high frame rate region, it is possible that the computational complexity can be reduced and the estimation accuracy can be improved, both due to the fact that the motion between two high frame-rate images is relatively smaller. It is therefore the purpose of this study to test this hypothesis and to evaluate performance of high frame-rate speckle tracking.ain factors affecting performance of high frame-rate speckle tracking include image signal-to-noise ratio (SNR), velocity and velocity gradient. Two sets of experiments and one set of simulations (bulk motion and motion with gradient) are conducted to evaluate the influences of the factors mentioned above. Speckle tracking using the sum-absolute-difference algorithm was adopted in this study. Plane-wave excitation was adopted to achieve high frame-rate imaging.he results for this method show that high accuracy can be achieved under sufficiently high SNRs. In addition, computational complexity of speckle tracking is reduced when frame rate is increased. Efficient and accurate high frame-rate speckle tracking is thus feasible.中文摘要…………………………………………………………… ibstract…………………………………………………………… iiiable of Contents………………………………………………… vist of Figures…………………………………………………… viiist of Tables……………………………………………………… xHAPTER 1 Introduction……………………………………… 1.1 Medical Ultrasound Imaging System……………………… 1.2 Development Trends…………………………………………. 2.2.1 High Frame-Rate Imaging……………………………. 3.2.2 Motion Vector Estimation……………………………. 4.3 Motivation and Aim……………………………………… 7.4 Thesis Framework……………………………………… 8HAPTER 2 High Frame-Rate Ultrasound Imaging...……. 10.1 Basic Beamforming……………………………………… 10.2 Conventional Focused imaging……………………………. 13.3 Multiple Line Transmission……………………………… 14.4 Parallel Receive Beamforming………………………… 15.4.1 Sparse Synthetic Aperture Imaging………………… 16.4.2 Plane-Wave Excitation Imaging…………………… 18.5 Imaging Quality of Research……………………………… 19HAPTER 3 Principles and Technique for Speckle Tracking 22.1 Origin of Speckle………………………………………….. 22.2 Speckle Decorrelation……………………………………… 25.3 Speckle Tracking Technique.……………………………… 273.3.1 Pattern Matching Algorithms………………… 29.3.2 Computational Complexity…………………………… 313.3.3 Error Analysis..………………………………… 33.4 Performance Evaluation Methods.......………………… 363.4.1 Tissue-Mimicking Phantom Experiment……… 373.4.2 Laminar Flow Simulation……………………… 383.4.3 Flow Experiment………………………………… 39.5 Displacement Estimation Process………………………… 40HAPTER 4 Performance Assessments……………………… 43.1 Influence of SNR………………………………………… 43.2 Influence of Uniform Velocity………………………… 45.3 Influence of Non-uniform Velocity…………………… 52.4 Discussion…………………………………………………… 58.5 Limitation of High Frame-Rate Speckle Tracking……… 61HAPTER 5 Concluding Remarks…………………………….... 63.1 Summary……………………………………………………… 63.2 Conclusions…………………………………………………… 63.3 Future Works………………………………………………… 64.3.1 Technique Improvement……………………………… 64.3.2 Clinical Applications…………………………………… 64eferences…………………………………………………………… 6