Integrating Segmentation and Neuro-Fuzzy Analysis for Efficient Glioma Brain Tumor Prediction

This paper introduces an innovative approach for the efficient prediction of glioma brain tumors by combining segmentation, Co-Active Adaptive Neuro Fuzzy Expert System (CANFES), and Principal Component Analysis (PCA). The CANFES model facilitates adaptive and dynamic learning, enabling the capture of complex patterns within the data for precise tumor classification. Experimental results highlight the effectiveness of the proposed method, showcasing significant improvements in sensitivity, specificity, and other performance metrics. Tumor segmentation results were achieved using the proposed methodology across various parameters. By implementing the PCA + CANFES approach, the study attained impressive results, including a precision rate of 98.85%, accuracy of 98.73 %, specificity of 98.56%, and sensitivity of 97.60%, 17.94 HD(cm), 0.911 k and 0.867 Cohen's kappa. Comparative analysis reveals that the PCA + CANFES method outperforms conventional approaches, demonstrating its potential for enhanced glioma tumor prediction.