Ultrasound image processing based on machine learning for the fully automatic evaluation of the Carotid Intima-Media Thickness

Atherosclerosis is responsible for a large proportion of cardiovascular diseases (CVD), which are the leading cause of death in the world. The atherosclerotic process, mainly affecting the medium- and large-size arteries, is a degenerative condition that causes thickening and the reduction of elasticity in the blood vessels. The Intima-Media Thickness (IMT) of the Common Carotid Artery (CCA) is a reliable early indicator of atherosclerosis. Usually, it is manually measured by marking pairs of points on a B-mode ultrasound scan image of the CCA. This paper proposes an automatic image segmentation procedure for the measurement of the IMT, avoiding the user dependence and the inter-rater variability. In particular, Radial Basis Function (RBF) Networks are designed and trained by means of the Optimally Pruned-Extreme Learning Machine (OP-ELM) algorithm to classify pixels from a given ultrasound image, allowing the extraction of IMT boundaries. The suggested approach has been validated on a set of 25 ultrasound images by comparing the automatic segmentations with manual tracings.