An Interactive Segmentation of Medical Image Series

In this paper, an algorithm based on the combination of the Canny operator and the morphology method is proposed for the semiautomatic segmentation of medical image series. Firstly, Canny operator is used to extract the whole accurate edges in the medical image series. Then find some object edge with the user interaction. And obtain the closed object edge by using the morphology methods of end point extraction, searching breakpoint, connecting breakpoints, removing burr and so on. Next, carry out expansion for the current object edge and make the expansion result as the location of object contour in the adjacent slice. With the same morphology methods, the closed object edge in the adjacent slice could be obtained automatically. Finally, make some interactive modification for the medical image series. Experiment shows that this algorithm can obtain the boundary of the desired object from a series of medical image quickly and reliably with only little user intervention.