MRI Brain Tumor Image Segmentation Using Region-Based Active Contour Model

Based on the brain tumor statistics provided by American Brain Tumor Association, the five-year survival rate for an adult who is diagnosed with malignant brain tumor is about 30 percent. It is estimated that new cases and deaths of brain tumor will reach 22020 and 13140, respectively in 2010. Early detection and diagnosis had been proven in increasing the chance of survival. Also, the medical centers had witnessed that the significance role of image segmentation as quantitative analysis in the diagnosis of diseases. Therefore, in order to accurately segment out and measure the size of brain tumor from MRI images, an algorithm for brain tumor segmentation using region-based active contour and preprocessing techniques is proposed. The results are tested with six images and compared with other method. Results from comparison show that the proposed method is more accurate and low computational cost than the other method. Key-Words: brain tumor, image segmentation, diagnosis, active contour, preprocessing, accurate