Sparsity-based Super-resolution Correlations Ultrasound Imaging

Inspired by super-localization methods in fluorescence microscopy, several methods for achieving super-localization in contrast-enhanced ultrasound (CEUS) imaging were presented in the last few years [1], [2], [3]. These methods enable the depiction of blood vessel structure and the flow within them, deep within the body, with unprecedented spatial resolution. However, all CEUS super-localization methods are currently limited by long acquisition times, impeding their application for clinical imaging. Recently, a statistical post-processing method named super-resolution fluctuation imaging (SOFI) [4] was applied to CEUS data and produced enhanced spatial resolution with very short acquisition times, using high ultrasound contrast agent concentrations [5]. In this work, we combine statistical tools with additional structural priors in the form of sparsity, to further improve the obtained resolution gain while maintaining high temporal resolution. The overall method is called Sparsity-based Ultrasound Super-resolution Imaging (SUShI).