RF signal local compression computation for imaging strains within a vessel mimicking cryogel phantom and a carotid artery

摘要

A major complication of atherosclerosis remains plaque ruptures which consequences can be acute thrombosis, myocardial infarct and sudden ischaemic death. Medium elastic properties are deeply involved in rupture mechanisms and can be investigated by Elastography, which estimates from pairs of RF ultrasound signals, the strain distribution occurring within a medium, in response to a stress. In this paper, elastograms obtained with an original 2-layer vessel mimicking cryogel phantom and a fresh excised human carotid artery are presented. The experimental setup consists mainly in a CVIS ultrasound scanner with a 30 MHz mechanical rotating single element, an oscilloscope Lecroy 9374L and a self-made pressuring system modifying the static lumen pressure by varying the fluid volume. RF data were digitised at a sampling frequency of 500 MHz, stored on disc and processed off-line with a method we developed, based on the principle that tissue deformation induces within RF signals, variations comparable to local compression factors. This method has been proved to be accurate for strains up to 7% and very adapted for investigating highly heterogeneous tissues like atherosclerotic arteries. Results have shown the method ability to separate regions with different mechanical properties, whereas the acoustical features are similar. Moreover the phantom elastogram exhibits between the two layers a strain ratio that matches perfectly reported abacus on the cryogel stiffness. In parallel arterial elastograms by presenting estimated strains in a wider range [0%-3.5%] than previous reported results, improve the discrimination between healthy and diseased regions.