Automatic Evaluation of Carotid Intima-Media Thickness in Ultrasounds Using Machine Learning

Cardiovascular diseases (CVD) are the main cause of death and disability in the world. Atherosclerosis is responsible for a large proportion of cardiovascular diseases. The atherosclerotic process is a degenerative condition, mainly affecting the medium- and large-size arteries, that develops over many years. It causes thickening and the reduction of elasticity in the blood vessels. An early diagnosis of this condition is crucial to prevent patients from suffering more serious pathologies. The evaluation of the Intima-Media Thickness (IMT) of the Common Carotid Artery (CCA) in B-mode ultrasound images is considered the most useful tool for the investigation of preclinical atherosclerosis. This paper proposes an effective image segmentation procedure for the measurement of the IMT in an automatic way, avoiding the user dependence and the inter-rater variability. Segmentation is raised as a pattern recognition problem and a neural network ensemble has been trained to classify the image pixels. The suggested approach is tested on a set of 25 ultrasound images and its validation is performed by comparing the automatic segmentations with manual tracings.