An Explainable Hybrid CNN–Transformer Framework for Accurate Brain Tumor Classification Using MRI Images

Abstract

Brain tumors are among the most life-threatening neurological disorders affecting human health worldwide. Early and accurate diagnosis plays a significant role in improving patient survival rates and treatment planning. Traditional manual diagnosis using Magnetic Resonance Imaging (MRI) is time-consuming and highly dependent on radiologists’ expertise. In recent years, Artificial Intelligence (AI) and Deep Learning techniques have demonstrated remarkable performance in medical image analysis. This research proposes an Explainable Hybrid Convolutional Neural Network (CNN) and Vision Transformer (ViT) framework for accurate brain tumor detection and classification using MRI images. The proposed model combines the local feature extraction capability of CNN with the global attention mechanism of Transformers to enhance classification accuracy. Furthermore, Explainable Artificial Intelligence (XAI) techniques such as Gradient-weighted Class Activation Mapping (Grad-CAM) are incorporated to improve model interpretability and assist medical practitioners in understanding tumor localization. The proposed methodology is evaluated using benchmark MRI datasets including BraTS and Kaggle Brain MRI datasets. Experimental results demonstrate that the hybrid framework achieves superior performance in terms of accuracy, precision, recall, F1-score, and computational efficiency compared to conventional deep learning models. The proposed system can contribute significantly toward intelligent healthcare systems and automated clinical decision support applications.