A Novel Hybrid Brain Tumor Segmentation Approach Using MOOT

The prompt and accurate detection of brain tumors is essential to ensure effective treatment and enhance the overall well-being of patients. This research presents an innovative approach for locating tumors in MRI scans. A Median-Anisotropic filter is proposed in the image preprocessing stage to eliminate noise while preserving edges. In the segmentation process, the Morphological Opening-Assisted Otsu Threshold (MOOT)method is proposed to enhance the precision of tumor localization. To extract the features of the tumor region the gray-level co-occurrence matrix (GLCM) is utilized, and finally, categorization of tumors was done by using a Random Forest classifier. The proposed method exhibits superior performance compared to existing approaches, highlighting its efficacy in tumor detection. This experimentation utilizes 600 brain tumor MRI images from the publicly available Figshare dataset. Training involves 60% of the images, constituting 360 images, while the remaining 40%, equivalent to 240 images, are reserved for testing. The experimentation achieves an accuracy of 97.92%.