Implementation of a Machine Learning-based Model for Cardiovascular Disease Post Exposure prophylaxis

Abstract

According to research, characteristics taken from ultrasound imaging may help with atherosclerosis diagnosis and decision-making. Atherosclerosis is estimated by parameters like elasticity, stiffness, lumen diameter, distension and IMT which can be used as an indicator of cardiovascular disease. Experienced radiologists are required to measure these parameters from the ultrasound images for the correct diagnosis. If a system could be automated to measure these parameters and thereby diagnose the CVDs, no doubt that this would be a milestone in the efforts taken to prevent cardiovascular disease. With the use of machine learning techniques, the classification of CCA anomalies from longitudinal ultrasonography B-mode images is progressed in this research. Acquisition of CCA images is done by ultrasound machine. For the segmentation of the layers of CCA, thresholding and other edge detection methods are used in spatial domain. Statistical, size and shape measurements are done from normal and abnormal images and features are identified.