Medical Image Segmentation Based On Level Set Method And Interactive Model

The virtual surgery is a fast developing area nowadays, and image segmentation is a classical question of image processing and analysis. Medical image segmentation plays a crucial role in the virtual surgery system, because it is the essential step of medical image processing. The automatic segmentation cannot satisfy the veracity of the medical image segmentation due to its complexity. Therefore the interactive segmentation method gets more and more attention, which is useful in this problem. This paper aims to do some research on interactive segmentation and deformable model, especially the geometric deformable model.There are basically two types of deformable models: parametric deformable model and geometric deformable model. Researching on this two models, and have a comparison, we find the latter is more fit for the medical image segmentation, because of the geometric deformable model, which is based on curve evolution theory and level set, can treat with the change of topology naturally. Level set is an effective method for curve or surface evolution in two spatial dimensions or three spatial dimensions. This paper do some research on the principle of the geometric deformable model level set, and introduce the narrow band method.Using the deformable model in image segmentation, we must choose a segmentation model, in order to form the energy field attracting the evolving curve to the boundary of our desired object. This paper introduces Mumford-Shah model and the C-V model, which is an improved M-S model. A new segmentation model, which combines the M-S model and narrow band scheme, is used to avoid computationally-time consuming. Contrast experiments between the M-S model andthe new model show that the latter decreases the time.In order to solve the problem that the veracity of automatic segmentation is correspondingly difficult, in some medical images, which are not enough clear, a restriction mechanism is presented based on level set method, and combined it with the narrow band M-S model. Then doctors are only needed to put few mark on the suitable image positions, and able to monitor the segmentation result. Experiments show that the model is reliable and practical.