Full field elastography using deep learning approach

Throughout the history of medicine, assessing stiffness through palpation has served as an indicator to gauge tissue health. Within our research team, we are advancing an innovative approach for full-field optical elastography, rooted in noise correlation analysis. This method leverages the relationship between the correlation function of a diffuse shear wave field and the time reversal of the shear wave field. By examining the correlation function, we then have access to an estimation of the shear wave speed, directly linked to tissue stiffness. Recent findings using this approach have shown great promise. However, in most cases, only the elasticity is quantified, despite the availability of additional information, such as viscosity, also present in the correlation function. In this paper, we introduce our initial outcomes in integrating noise correlation with artificial intelligence. More specifically, we employ a U-NET-based architecture to process noise correlation data.