Brain Tumor Detection from MRI Images Using Window-Aware Hierarchical Auto-Associative Polynomial Network with Great Wall Construction Algorithm

Early brain tumor (BT) diagnosis is plagued by a small number of samples, limited generalizability to other groups, and a lack of ability to detect subtle variation in dense brain tissue. A new technique called Window-Aware Hierarchical Auto-Associative Polynomial Network with the Great Wall Construction Algorithm (WA-HAAPNet-GWCA) has been developed to overcome these problems. This method uses Magnetic Resonance Imaging (MRI) images from the CE-MRI (Contrast-Enhanced Magnetic Resonance Imaging) and Figshare datasets for assessment. Window-Aware Guide Filtering (WAGF) efficiently lowers noise and artifacts during preprocessing, improving image quality. R2U + + with Tuning Attention (R2U-TA) is used to locate tumors precisely. While the Scale-Aware Hierarchical Auto-Associative Polynomial Network (SA-HAAPNet) is used for feature extraction and classification, the Great Wall Construction Algorithm (GWCA) enhances model performance and reliability. The WA-HAAPNet-GWCA approach is able to use the CE-MRI and Figshare datasets and obtains a recall of 99.8% and accuracy of 99.9%, highlighting a very high diagnostic accuracy. The approach is efficacious and shows the feasibility of clinical use by providing solid answers to problems that have eluded diagnosis. This strategy, with elements drawn from the cutting edge, ensures cutting-edge performance in uncovering BTs at an early stage, which significantly enhances patient outcomes and contributes to the development of precision medicine.