Automatic estimation of the arterial parameters in Ultrasound video sequence

In this article we propose a new system for the accurate estimation of the center and the radius of a circular object in a medical video sequence. The system takes as an input a region of interest (ROI) where the object exists, this area is given by an object detector. The contribution of the presented method is the high accuracy of estimating the position of both the center and radius of the artery. The results are comparable with other approaches and are less expensive computationally. The detection of a circular object may seem a classic task in image processing. However, because of the special nature of Ultrasound images, it becomes a harder task which needs a different procesing. The system proposed in this article is fully automatic, and its results showed a rate of 89% accuracy for the correct detection of the center and the radius of artery. The system uses circular hough transform as a final step to accurately detect the sought object. The system is tested on B-mode Ultrasound (US) images. The boundary of the common carotid artery (CCA) is extracted, and its radius is calculated. This is an initial step towards analysis of cardiac diseases, or for tracking the movement of CCA.