Algorithm Of Medical Image Segmentation And Compression

The medical digital image processing has a wide application in field of medical image and clinic. The research of medical image processing takes important role in the medical application, and medical image segmentation and compression technology have especially demand in PACS of digital hospital in recent years.Medical image processing technologies are introduced in this paper at first, and medical image segmentation and compression are studied. This paper focus on the following three points:1. The application of Markov Random Field (MRF) has been studied in this article,that based on particle swarm optimization (PSO) for magnetic resonance image (MRI) segmentation. Imagery segmentation of MRF-PSO model based on MRF is firstly proposed. Then the maximum a posteriori (MAP) global best solution of segmentations will be got though MRF by using the method of PSO, which describes image data relations by local correlations instead of global image possibility distributions. Finally, results are given. It shows that the MRF-PSO method is an effective method in image segmentation.2. Considering medical image feature, we bring forward a new thought of medical image segmentation based Data-Driver Markov chain model and carry it out. The Algorithm creates a Markov Chain which is a closed curve by shifting probability. To accelerate convergence, Monte Carlo Algorithm is applied into MCMC model. The experiments on representative databases indicate the validity of it. The experiment indicates the algorithm has high ability of anti-noise, and can achieve accurate medical image segmentation.3. Based on the characteristic of medical image that whose information relatively concentrate, an compound algorithm for ROI (Region of Interest) compression of medical image based on EBCOT(Embedded Block Coding with Optimized Truncation) and DWT is proposed. The ROI and the BC (background) are processed with different coding methods, which can assure a high compression ratio and a high quality of ROI in the medical image. Experiment results show that the method is an effective method in compression and image restruction.