Texture analysis of carotid plaque ultrasound images

Abstract

In this work, analysis of carotid plaque ultrasound images have been attempted using statistical method based on Gray Level Co-occurrence matrix (GLCM). The ultrasound imaging of the carotid arteries is a common study performed for diagnosis of carotid artery disease. The first order linear scaling filter was used to enhance the image quality for analysis. Second order statistical texture analysis is performed on the acquired images using GLCM method and a set of 12 features are derived. Principal Component Analysis (PCA) is employed to reduce features used for classifying normal and abnormal images to four which had maximum magnitude in the first principal component. It appears that, during diagnosis this method of texture analysis could be useful to develop an automated system for characterisation and classification of carotid ultrasound images.