Абстрактный

Common Carotid Artery Intima Media Thickness Segmentation

L. Ashok Kumar, K.Karthika, S.Kavitha, T. Vinoth Krishnan

Stroke is one of the leading threats to human health. Stroke and heart attack would seriously cause human morbidity and mortality. The carotid artery of intima-media thickness is a key indicator to the disease and is one of the major criteria for early screening and diagnosis of stroke. The intima-media thickness can be measured by segmenting the ultrasound images of carotid artery. First, the ultrasound image is loaded and the image is enhanced by reducing the speckle noise by using wavelet filter. Then the enhanced image is segmented into contours using level set method in which Chan-Vese algorithm is used. The level set formulation by Chan and Vese using random initialization provides a segmentation of the common carotid artery ultrasound images into different distinct regions. Finally, the resulting segmented image is classified into several classes which is used for IMT measurement of carotid artery. The resulting smoothed lumen intima boundary combined with anatomical information providean excellent initialization for parametric active contours that provide the final IMC segmentation which is used for intima-media thickness measurement.