A brain tumor is characterized by the abnormal proliferation of cells within the brain, forming irregular tissue masses. This condition poses serious health risks and can significantly diminish a patient’s quality of life. To enhance the accuracy of brain tumor detection and segmentation from MRI scans, we propose a novel approach that integrates Explainable Artificial Intelligence (XAI) with the Enhanced Fuzzy C-Means (ENFCM) algorithm. The proposed system not only accurately classifies various tumor types but also provides transparent insights into the model’s decision-making process. Experimental results demonstrate that the integration of XAI and ENFCM significantly improves both the accuracy and efficiency of tumor prediction and progression monitoring. Notably, our model achieves an accuracy of 98.63% on the Br35H dataset. XAI techniques such as Grad-CAM and SHAP further enhance the interpretability of predictions, while ENFCM improves segmentation robustness, particularly in low-contrast images. This comprehensive approach supports clinicians in diagnosis and treatment planning, while also laying the groundwork for future advances in AI-driven medical applications, ultimately contributing to improved outcomes for patients affected by brain tumors.