Automatic ultrasound scanning for the Carotid Artery by robotic system

Abstract

In this paper, we present an autonomous robotic system which is composed of a 7-DoF manipulator and a conventional medical ultrasound diagnostic equipment, which used to automatically search and detect the carotid artery for diagnostic purposes. In the acquired real-time images, multiple anatomical features are used to detect carotid artery landmarks through a yolov5-based detection model. Based on the image processing results, a scanning strategy was designed to search for the desired imaging plane of the carotid artery by adjusting the position and rotation of the ultrasound probe. The experimental results show that the system can achieve high accuracy carotid artery detection and more clarity ultrasound images by adjusting the position and attitude of the probe to achieve automatic scanning of the carotid artery.