ABSTRACT
Medical images are efficiently and effectively diagnosed using a variety of segmentation and classification techniques. Because the medical anatomy varies greatly, the MR image is segmented at the early stage of diagnosis due to a variance among numerous brain and image characteristics. The suggested technique extracts ideas by applying an algorithm that carries out a comparable picture segmentation procedure. A Modified Decision Based Coupled Windows Median Filter (MDBCWMF) is used in the filtering process. When compared with K-means and FCM, the acquired findings show a significant reduction in the segmentation error assessed using the Modified Distance Metric of FCM methods. By effectively diagnosing the majority of brain tumor pictures. The filtering method employed in this paper produced better results than median and PGPD. When the suggested segmentation parameters are applied to the brain imaging data, improved results are produced when compared to the ground truth images.
