Myocardial D-Spect Image Segmentation Using RSF Model and Kernal Difference Operator

One of the top causes of death worldwide is now ischemic heart disease. The modern diagnostic technology known as dynamic single-photon emission computed tomography (D-SPECT) is frequently utilized to confirm the myocardial function in patients with a variety of heart disorders. To build a 3-D Myocardial model and let doctors conduct assessments of cardiac function, accurate automatic localization and segmentation of myocardial areas is beneficial. Image segmentation is a crucial method in preclinical cardiac research because of this. One of the frequent problems with image segmentation is intensity inhomogeneity, which is brought on by imperfect instrumentation and image blemishes. This research presents a unique region-based active contour model that can precisely segment the myocardial D-SPECT image. As RSF Model detects the intensity inhomogeneity of the object and KD operator makes the weak boundaries of the object strong. By integrating the RSF (region-scalable fitting) energy model and the KD (kernel difference) operator, the suggested method ensured accuracy, computational efficiency where it showed results in less iterations and can address the problem of high sensitivity for Myocardial D-SPECT segmentation.