PROJECT TITLE :

Efficient Segmentation Methods for Tumor Detection in MRI Images - 2014

ABSTRACT:

Brain tumor extraction and its analysis are difficult tasks in medical image processing as a result of brain image and its structure is sophisticated which will be analyzed solely by expert radiologists. Segmentation plays an vital role within the processing of medical pictures. MRI (magnetic resonance imaging) has become a notably helpful medical diagnostic tool for diagnosis of brain and other medical images. This paper presents a comparative study of three segmentation ways implemented for tumor detection. The strategies include k-suggests that clustering with watershed segmentation algorithm, optimized k-suggests that clustering with genetic algorithm and optimized c- means that clustering with genetic algorithm. Ancient k-suggests that algorithm is sensitive to the initial cluster centers. Genetic c-suggests that and k-means that clustering techniques are used to detect tumor in MRI of brain images. At the end of process the tumor is extracted from the MR image and its precise position and the shape are determined. The experimental results indicate that genetic c-means not solely eliminate the over-segmentation downside, but also give quick and efficient clustering results.