Classification of plaque in ultrasonic images of human arteries

Abstract

Cardiovascular disease is a major health problem in the industrialised world. The most common underlying disease is the atherosclerosis that can lead to myocardial infarction, in the brain to cerebral infarction or stroke, and in the peripheral vascularutrearteries to gangrene and loss of function. Investigation of the carotid arteries by using ultrasound has become a predominant tool during the last decade. The parameter most frequently used in the clinical diagnosis today is the degree of plaque, determined by ultrasound duplex examination. In the process of developing a help tool for subjective visual classification, we introduce a study on the correlation between subjective visual classification and statistical and neural-network based classification on B-mode ultrasound images of atherosclerotic plaque. Since the subjective classification is a very difficult visual issue of categorising different types of plaque, it is easily solved by the computer help toll classification. Experiments conducted so far yield encouraging results.