Cross-spectral matrix fitting for passive mapping of the ultrasonic cavitation based on Elastic-net regularization

Cavitation produced by high-intensity focused ultrasound needs to be monitored for specific applications such as sonoporation. Passive acoustic mapping using adaptive or non-adaptive beamforming methods has been proposed in the literature for this task, but lacks of spatial (primarily axial) resolution. This is mainly due to the absence of a time reference, a common specificity of passive imaging compared to standard pulse-echo ultrasound. In this paper, we propose an inverse problem-based method for passive cavitation imaging, aiming to increase the spatial resolution through sparsity constraints. Results on simulated data are presented and compared to state-of-the-art approaches, such as the frequential delay-and-sum and the frequential robust Capon beamformer.