An Efficient Segmentation Approach for Brain Tumor Detection in MRI

Background: For effective tumor diagnosis, early brain tumor detection becomes an important procedure. Despite a huge number of tumor detection techniques available, brain tumor segmentation is still a challenging field because of the complex characteristic of the brain MR images. This work aims to achieve an efficient segmentation approach for tumor detection. Methods: The Contextual Clustering based segmentation methodology proposed here includes image pre- processing and tumor segmentation. Image pre-processing removes total noise in the image and corrects the boundaries. Tumor segmentation uses Contextual Clustering algorithm to segment the tumor part from the input MR images. Findings: An automatic method of tumor detection and localization in the brain MRI is proposed here which avoids false segmentation and improves accuracy. Application: This stated Contextual Clustering algorithm works efficiently in brain tumor segmentation for the MRI brain images and produces accurate results for the input datasets and used in medical fields.