Active contour-based segmentation of medical image series

In this paper, an algorithm for the semiautomatic segmentation of medical image series is proposed by combining the live wire algorithm and the active contour model. First, we use the robust anisotropic diffusion filtering to smooth the images while keeping the edges. Then we modify the traditional live wire algorithm by combining it with the watershed method. Using the improved live wire method, the accurate segmentation of one or more medical images could be obtained firstly. Based on the segmentation of previous slices, the computer will segment the nearby slices using the modified active contour model automatically. To make full use of the correlative information between contiguous slices, a gray-scale model is applied to the model to record the local region characters of the desired object, and a new functional definition of the external energy is designed. Furthermore, in order to be adaptable with the topological change of the nearby slices, affine cell image decomposition is applied to the active contour model. The experiment results show that this algorithm can recover the boundary of the desired object from a series of medical images quickly and reliably with only little user intervention.