An improved method for detection of carotid walls in ARTSENS

We have been developing a fully automated ultrasound based imageless system to facilitate mass screening of patients for future risk of cardiovascular diseases. The device shall enable a general medical practitioner to non-invasively measure the local arterial stiffness of common carotid artery (CCA) and has been acronymed ARTerial Stiffness Evaluation for Non-invasive Screening (ARTSENS™). Complete automation of the system requires providing assistance in placement of probe over the CCA location and automatic identification of approximate location of proximal wall (PW) and distal wall (DW) of the CCA. In this paper we propose a method based on temporal motion of PW and DW over successive A-Mode frames to locate the CCA. We evaluated the performance of the algorithm with data obtained from CCA of 30 subjects. It could correctly identify the CCA in more than 70 % of trials. We also propose a method for preprocessing the frames by using the transmitted pulse wavelet. This improved the detection rate significantly. False positives were always less than 6% of total detections.