An easily-achieved time-domain beamformer for ultrafast ultrasound imaging based on compressive sensing.

Abstract

In ultrafast ultrasound imaging technique, how to maintain the high frame rate, and at the same time to improve the image quality as far as possible, has become a significant issue. Several novel beamforming methods based on compressive sensing (CS) theory have been proposed in previous literatures, but all have their own limitations, such as the excessively large memory consumption and the errors caused by the short-time discrete Fourier transform (STDFT). In this study, a novel CS-based time-domain beamformer for plane-wave ultrasound imaging is proposed and its image quality has been verified to be better than the traditional DAS method and even the popular coherent compounding method on several simulated phantoms. Comparing to the existing CS method, the memory consumption of our method is significantly reduced since the encoding matrix can be sparse-expressed. In addition, the time-delay calculations of the echo signals are directly accomplished in time-domain with a dictionary concept, avoiding the errors induced by the short-time Fourier translation calculation in those frequency-domain methods. The proposed method can be easily implemented on some low-cost hardware platforms, and can obtain ultrasound images with both high frame rate and good image quality, which make it has a great potential for clinical application.