Weighted Fuzzy C Means: A Novel Tumor Segmentation Approach in MR Brain Images

The most prevalent primary brain tumor, glioma, is caused by glial cell carcinogenesis in the central nervous system. For numerous applications in the area of health care evaluation, brain tumor localization and separation from magnetic resonance images (MRI) are challenging yet crucial tasks. Several recently developed methods utilized four modalities: T1, T1c, T2, and FLAIR. This is because each brain imaging modality provides distinct and important information concerning every region of the tumor. The process of diagnosis, therapy selection, and risk variables detection depends on trustworthy and precise tumor segmentation and survival of patients forecasting. In this article, a state-of-the-art fuzzy-based system is introduced that uses multimodal MRI images to categorize brain tumors and estimate glioma survival. To address the drawbacks of the FCM, the suggested approach combined the weight function with the conventional Fuzzy C-means (FCM). Extensive tests are carried out on the different BRATS challenge datasets, demonstrating that the suggested approach achieves competitive outcomes. Evaluation on the BraTS dataset confirms the effectiveness of the developed Weighted FCM (WFCM), and the segmented results are compared with the ground truth images. A small number of performance metrics were also used for assessing the qualitative as well as quantitative outcomes. The resulting dissected images help medical professionals diagnose, medicate, or plan for medical intervention for the affected individuals earlier.