Medical image segmentation by geodesic active contour incorporating region statistical information

Geodesic active contour methods are powerful numerical techniques for image segmentation and analysis. However the edge detector in the model ensures that the data on both sides of the contour is as dissimilar as possible, it cannot deal with the requirement that also the interior of a region should be as homogeneous as possible. In this paper, we present a new region based geodesic active contour method. The new method gives a global view of the boundary information within the image. The method here proposed is particularly well adapted to situations where edges are weak and overlap, and the curve initialization is a very bad guess. A number of experiments on CT medical images were performed to evaluate the new method. The experimental results demonstrate the reliability and efficiency of this new method.