Fast pulse-echo ultrasound imaging employing compressive sensing

In fast pulse-echo diagnostic ultrasound imaging the acquisition time for a single image is crucial. This time depends significantly on the number of sequentially emitted sound waves. For fewer wave emissions the inverse scattering problem is increasingly ill-posed. In this contribution we establish and investigate a solution based on compressive sensing (CS). This approach accounts for the lack of measurement data by assuming sparsity of the material parameters in an arbitrary basis. Using measurements obtained from a wire and a multi-tissue phantom, we evaluate the performance of our CS solution in comparison to synthetic aperture focussing, delay-and-sum (DAS) beamforming and filtered backpropagation (FBP). Emitting only a single plane wave, the CS approach yields the best results for the sparse wire phantom in terms of sidelobe reduction and lateral -6 dB-widths. For the non-sparse multi-tissue phantom, we observe equivalent results for CS, DAS, and FBP when a single plane wave is emitted and a discrete cosine basis is employed as sparsifying transform.