The Research Of The Brain Tumor Sequence Images Segmentation Based On The MRI

MRI is widely used in the diagnosis of soft tissue due to its unique characteristic of noninvasinveness. MRI segmentation is the essential step of MRI processing, and it plays a crucial role in both qualitative and quantitive MRI analyses. Currently, the most widely used segmentation approach in a clinical MRI systerm is based on manual delineation. To accomplish MRI segmentation more efficiently and accurately, a computerized approach would be an ideal choice for clinical use. A computerized approach is expected to segment the object of interest automatically or semi-automatically with high reproducibitity.Active contour model is widely used in segmentation of medical image due to it can combine the high-level vision knowledge and low-level image information. This paper puts forward a new medical sequence image segmentation algorithm based on Snake and the greedy algorithm. This paper is designed to exactly segment the brain tumor from the brain MRI, automatically and accurately. Firstly the Anisotropic Diffusion is used for image preprocessing. Then by analyzing the original active contour model and some main modify models, the internal energy and external energy of active contour model are modified as follows:1.To avoid the convergence of neighbor control point, the average contour length term is added into the internal energy of the model.2.To avoid the active contour find wrong boundary, the gradient directional energy is introduced to the external energy of the model.3.The region energy, based on characteristic of image is regarded as the external energy, can overcome the influence of noise. Then the greedy algorithm is used to find the minimum energy of the model. A fast algorithm is introduced to solve the minimum of region energy. The algorithm of add or delete snaxel also adopted in this paper. And proposed an initial contour selection of snake model.To improve the automation of the sequence images segmentation, first we should find out a key image and divide the whole sequence picture into two part, then segment the key accurately. Because the sequence images become gradually, the last segmentation can predict the initial contour of the next image, then accomplish automatic segmentation of the brain tumor serial images.Segmentation of the MRI brain tumor are studied in the experiments. Comparing to the manual segmentation and the GVF segmentation. The segmentation results of the proposed model is approximate to the manual delineation and better than GVF. Segmentation results of different shape and size of initial contour proved that the model is none sensitive to initial contour. Gray statistical characteristics are used to automatically initiate Snake in the key image, So the algorithm rose the automatic degree of the Serial Images Segmentation. The results of experiments indicate that the proposed segmentation method can achieve excellent segmentation results.