An Autocorrelation-Based Approach for Estimating Shear Wave Speed using 2D Velocity Vector

Estimating shear wave speed in human tissues provides valuable insights for diagnosing various diseases. These estimates are used, for instance, to determine the stage of certain diseases. The implementation of more accurate and robust data processing methods has the potential to increase the reliability of elastography imaging. In this paper, we propose an algorithm for time-harmonic elastography that utilizes a two-dimensional velocity vector as opposed to using only one dimension. The estimated velocity vectors are projected onto a line that is perpendicular to the shear wave propagation direction. After directional filtering along different angles, the shear wave speed was estimated using a spatial autocorrelation method.We compared the proposed algorithm with the phase-gradient approach using synthetic data generated from k-Wave simulations of a speckle phantom with a stiff inclusion. The results showed that the proposed method has a 49% and 73% lower bias in the inclusion and background regions, respectively. The proposed method shows promising results with lower variance and bias; and further evaluation is underway using in vivo and phantom recording data.