Ultrafast Ultrasound Speckle Tracking Using the Dynamical Kernels for Improved Blood Flow Velocity Profiling

The accurate measurement of blood flow velocities is of great importance for the prevention, early diagnosis, and course monitoring of atherosclerosis. Unlike the conventional speckle tracking using the pre-defined kernels (STPK), an improved speckle tracking using the dynamical kernels (STDK) based on ultrafast ultrasound is proposed to improve the accuracy and efficiency of the blood flow velocity profile (BFVP) estimation in this work. The novelty of this study is that the coordinate of the current tracked kernel in the reference frame is dynamically adjusted according to the coordinate of the best matching of the last tracked kernel in the comparison frame. This modification is not only fault-tolerant to a certain degree for speckle decorrelation, but also reduces the time consumption produced by traversing the whole search regions. The performance of the proposed STDK was validated via simulations. Compared with the STPK-based results, the NRMSE and acquisition time of the STDK-based results decrease by 48.39% and 36.8% on average, respectively. Therefore, the proposed STDK method has been proven to improve the accuracy, efficiency of the BFVP estimation, which is helpful for clinical diagnosis and treatment of related diseases.