Robust estimation of carotid artery wall motion using the elasticity-based state-space approach.

摘要: Abstract The dynamics of the carotid artery wall has been recognized as a valuable indicator to evaluate status atherosclerotic disease in preclinical stage. However, it is still challenge accurately measure this from ultrasound images. This paper aims at developing an elasticity-based state-space approach for measuring two-dimensional motion imaging sequences. In our approach, we have employed linear elasticity model wall, and converted into state space equation. Then, computed by solving using H∞ filter block matching method. addition, parameter training strategy proposed study dealing with initialization problem. experiment, also developed evaluation function tracking accuracy considering influence sizes two blocks (acquired manual tracing) containing same tissue their overlapping degree. compared performance traced results drawn three medical physicians on 37 healthy subjects 103 unhealthy subjects. showed that was highly correlated (Pearson’s correlation coefficient equals 0.9897 radial 0.9536 longitudinal motion), agreed well (width 95% confidence interval 89.62 µm 387.26 µm motion) tracing We kinds previous methods, including conventional Kalman-based methods optical flow. Altogether, able successfully demonstrate efficacy elasticity-model based (EBS) more accurate 2-dimensional towards effective assessment