Recovery of compressively sensed ultrasound images with structured Sparse Bayesian Learning

In this paper, we consider the problem of recovering compressively sensed ultrasound images. We build on prior work, and consider a number of existing approaches that we consider to be the state-of-the-art. The methods we consider take advantage of a number of assumptions on the signals including those of temporal and spatial correlation, block structure, prior knowledge of the support, and non-Gaussianity. We conduct a series of intensive tests to quantify the performance of these methods. We find that by altering the parameters of the structured Sparse Bayesian Learning approaches considered, we can significantly improve the objective quality of the reconstructed images. The results we achieve are a significant improvement upon previously proposed reconstruction techniques. In addition, we further show that by careful choice of parameters, we can obtain near-optimal results whilst requiring only a small fraction of the computational time needed for the best reconstruction quality.