Greedy Active Contour Detection for Median Nerve on Strain Sonographic Images

Carpal tunnel syndrome (CTS) is commonly occurred in occupations using vibrating manual tools or handling tasks with highly repetitive and forceful manual exertion. Recently, the ultrasonography has been used to evaluate CTS by monitoring median nerve movements. In order to facilitate the automatic extraction of shape characteristics for the median nerve, this paper designed a procedure that used greedy active contour detection model (GACD) to detect the edge of median nerve in ultrasound image. We selected a ROI to be an initial of virtual contour for median nerve in original ultrasound image. That can enhance the sensitivity of proposed GACD model to detect the contour of median nerve. In the experiment, the results show that the performance of the method is feasible and accurate.