Medical Image Classification and Recognition Based on Fuzzy Statistical Coupling Sub-Histogram

Magnetic resonance imaging (MRI) plays an important role in the medical field and can provide the qualitative, quantitative and accurate information for medical diagnosis. MRI is more advantageous than other modes of medical imaging technology in many application fields and meanwhile, the accurate and robust brain MR image segmentation, feature extraction and classification are very important for clinical cancer diagnosis. A new cancer diagnosis method for brain MR image is proposed. The adoption of MR image to extract brain tumor GMM feature algorithm is proposed and realized. The technologies such as multi-threshold segmentation, brain tumor GMM feature calculation and extraction and the discrimination between brain tumor and normal part based on decision tree classifier are used in brain tumor identification or diagnosis. MR image simulation results based on brain tumor test proves that the extraction technology by GMM feature obtains the expected results. The preliminary experimental results indicate that the method performs well in cancer cell and brain tumor test. The capacity of the method is reflected in its higher precision than other algorithms in brain tumor test.