Adaptive beamformer with echo statistics

Abstract

Delay-and-sum (DAS) beamforming is commonly used in commercial scanners and computationally efficient for real-time imaging. However, the ability to suppress off-axis signals is limited. A minimum variance (MV) beamformer realizes a superior performance in suppression of off-axis signals and improves lateral resolution significantly. On the other hand, MV degrades the contrast-to-noise ratio (CNR) compared to DAS because it alters speckle statistics. In this study, we developed a method for improvement of CNR in MV beamforming by evaluating envelope statistics of echo signals. It is well known that the envelope statistic of speckle echoes from a random medium obeys the Rayleigh distribution. The echo envelope statistics were evaluated using the shape parameter of the Nakagami distribution. The proposed beamformer is worked as DAS for speckle echoes and MV for non-speckle echoes by referring to the Nakagami shape parameter. In the phantom experiment, the lateral resolution of MV was 0.18 mm, which was significantly better than 0.53 mm obtained by DAS. However, CNR was degraded from 6.73 dB to 4.05 dB by MV. The proposed beamformer realized a lateral resolution of 0.25 mm, which was significantly better than DAS, with a CNR value of 6.03 dB, which was comparable to DAS.