An Active Contour Model Based Multi Objective Optimization Technique for Coronary Calcium Scoring

Developments in healthcare technologies have significantly enhanced spatial resolution, facilitating better anatomical elements and improved contrast resolution, permitting analysis of additional subtle structures than formerly attainable. Automatic recognition and quantification of calcifications from arteries in computed tomography (CT) scans is a key necessity in planning the treatment of individuals with suspected coronary artery disease. A method for automated coronary artery segmentation has been proposed for the quantification of calcium objects. Initially, morphological functions like erosion and dilation are used to enhance the vessels, following which a multi-phase, multi-objective optimization technique based on the active contour model has been proposed for coronary artery segmentation. After segmenting the arteries, coronary calcium objects are detected, based on intensity and size features. Coronary CTs obtained from the Somatom Definition AS + CT scanner with a slice thickness of 3 mm was obtained from clinical practice. Experimental results demonstrate that our proposed method provides optimal segmentation of coronary arteries so as to improve the accuracy of coronary calcification detection.