A motion adaptable needle placement instrument based on tumor specific ultrasonic image segmentation

Abstract

This paper suggests an ultrasound guided needle insertion instrument which can track target motion in real-time. Under traditional ultrasound guided needle insertion therapies, surgeons have had much burden to find out the precise targeting position, particularly when the organ is moving due to the respiration or heartbeat. We developed a new needle insertion instrument which can track moving target based on visual servo control. In addition, this paper proposed a tumor specific active contour model which can conduct a fast and robust segmentation for tumor, and utilized Hough transform for needle recognition. In the experiment, the proposed system could track a moving phantom successfully at speed of 3 frames/sec processing.