Abstract-Conventional ultrasound (US) image reconstruction methods rely on delay-and-sum (DAS) beamforming, which is a relatively poor solution of the image reconstruction problem. An alternative to DAS consists in using iterative techniques which require both an accurate measurement model and a strong prior on the image under scrutiny. Towards this goal, much effort has been deployed in formulating models for US imaging which usually require a large amount of memory to store the matrix coefficients. We present two different techniques which take advantage of fast and matrix-free formulations derived for the measurement model and its adjoint, and rely on sparsity of US images in well-chosen models. Sparse regularization is used for enhanced image reconstruction. Compressed beamforming exploits the compressed sensing framework to restore high quality images from fewer raw-data than state-of-the-art approaches. Using simulated data and in vivo experimental acquisitions, we show that the proposed approach is three orders of magnitude faster than non-DAS state-of-the-art methods, with comparable or better image quality.

No abstract

No abstract

Abstract-Based on the success of deep neural networks for image recovery, we propose a new paradigm for the compression and decompression of ultrasound (US) signals which relies on stacked denoising autoencoders. The first layer of the network is used to compress the signals and the remaining layers perform the reconstruction. We train the network on simulated US signals and evaluate its quality on images of the publicly available PICMUS dataset. We demonstrate that such a simple architecture outperforms state-of-the art methods, based on the compressed sensing framework, both in terms of image quality and computational complexity.

Pulse-echo ultrasound (US) aims at imaging tissue using an array of piezoelectric elements by transmitting short US pulses and receiving backscattered echoes. Conventional US imaging relies on delay-and-sum (DAS) beamforming which retrieves a radio-frequency (RF) image, a blurred estimate of the tissue reflectivity function (TRF). To address the problem of the blur induced by the DAS, deconvolution techniques have been extensively studied as a post-processing tool for improving the resolution. Most approaches assume the blur to be spatially invariant, i.e. stationary, across the imaging domain. However, due to physical effects related to the propagation, the blur is nonstationary across the imaging domain. In this work, we propose a continuous-domain formulation of a model which accounts for the diffraction effects related to the propagation. We define a PSF operator as a sequential application of the forward and adjoint operators associated with this model, under some specific assumptions that we precise. Taking into account this sequential structure, we exploit efficient formulations of the operators in the discrete domain and provide a PSF operator which exhibits linear complexity with respect to the grid size. We use the proposed model in a maximum-a-posteriori estimation algorithm, with a generalized Gaussian distribution prior for the TRF. Through simulations and in-vivo experimental data, we demonstrate its superiority against state-of-the-art deconvolution methods based on a stationary PSF.