We propose an approach to quantify myocardial viability from late Gadolinium enhancement Cardiac Magnetic Resonance (LGE-CMR) images in patients with ischemic cardiomyopathy. After manual delineation of myocardial borders, segmentation of myocardial scar tissue is performed automatically by applying a classifier based on Gaussian mixture model (GMM) of the myocardium intensity histogram. Segmentation result is refined based on the a priori knowledge of the hyper-enhanced pattern of LGE-CMR in

more »

... chemic patients, and a dedicated parametric dense representation of scar transmurality is proposed. High level of accuracy and correlation against the manual reference technique envisage the clinical applicability of automated processing technique, minimizing user interaction and the time required for the analysis.