An Explainable AI System for Efficient Brain Tumor Diagnosis on Edge Devices using Transfer Learning

  The proposed Explainable Artificial Intelligence (XAI)–based Brain Tumor Diagnosis System effectively combines deep learning, image enhancement techniques, and explainable methods to deliver accurate, transparent, and efficient tumor detection from MRI scans. This framework addresses key challenges in modern neuro-diagnostic practices, including limited availability of expert clinicians, computational limitations in resource-constrained settings, and the lack of interpretability associated with conventional deep learning models. By employing advanced preprocessing approaches such as Independent Component Analysis (ICA), Adaptive Filtering, and Contrast Limited Adaptive Histogram Equalization (CLAHE), the system improves MRI image quality and enhances tumor visibility. Lightweight transfer learning architectures, including MobileNet, EfficientNet, and Tiny Vision Transformers (ViT), achieve strong diagnostic accuracy while maintaining computational efficiency suitable for real-time implementation. A significant contribution of this study is the integration of explainability techniques such as Grad-CAM, SHAP, and LIME, which provide both visual and quantitative explanations for model predictions, thereby increasing clinical reliability and enabling radiologists to validate the decision-making process. Additionally, the incorporation of tumor segmentation, size estimation, and growth monitoring supports effective treatment planning and patient management. In summary, the developed system fulfills its core objectives by delivering an accurate, interpretable, and resource-efficient solution for brain tumor detection and analysis. It narrows the gap between artificial intelligence technologies and clinical practice by offering a transparent and deployable model suitable for real-world healthcare applications. Future developments may include integrating multimodal medical datasets such as CT, PET, and genomic information, adopting federated learning techniques for secure data sharing, and implementing cloud–edge hybrid frameworks to enable large-scale clinical deployment. These enhancements are expected to expand the system’s role in personalized medicine and intelligent healthcare systems. 

[1] S. Solanki, U. P. Singh, S. S. Chouhan and S. Jain, "Brain Tumor Detection and Classification Using Intelligence Techniques: An Overview," in IEEE Access, vol. 11, pp. 12870-12886, 2023, doi: 10.1109/ACCESS.2023.3242666 [2] K. Lata, P. Singh, S. Saini and L. R. Cenkeramaddi, "Deep Learning-Based Brain Tumor Detection in Privacy-Preserving Smart Health Care Systems," in IEEE Access, vol. 12, pp. 140722-140733, 2024, doi: 10.1109/ACCESS.2024.3456599 [3] A. A. Asiri, T. A. Soomro, A. A. Shah, G. Pogrebna, M. Irfan and S. Alqahtani, "Optimized Brain Tumor Detection: A Dual-Module Approach for MRI Image Enhancement and Tumor Classification," in IEEE Access, vol. 12, pp. 42868-42887, 2024, doi: 10.1109/ACCESS.2024.3379136. [4] H. A. Shah, F. Saeed, S. Yun, J. -H. Park, A. Paul and J. -M. Kang, "A Robust Approach for Brain Tumor Detection in Magnetic Resonance Images Using Finetuned EfficientNet," in IEEE Access, vol. 10, pp. 65426-65438, 2022, doi: 10.1109/ACCESS.2022.3184113. [5] A. Younis et al., "Abnormal Brain Tumors Classification Using ResNet50 and Its Comprehensive Evaluation," in IEEE Access, vol. 12, pp. 78843-78853, 2024, doi: 10.1109/ACCESS.2024.3403902. [6] Amin, J., Sharif, M., Haldorai, A. et al. Brain tumor detection and classification using machine learning: a comprehensive survey. Complex Intell. Syst. 8, 3161–3183 (2022). https://doi.org/10.1007/s40747-021-00563-y. [7] Hosny, K.M., Mohammed, M.A. Explainable AI and vision transformers for detection and classification of brain tumor: a comprehensive survey. Artif Intell Rev 58, 259 (2025). https://doi.org/10.1007/s10462-025-11221-x [8]  Mathivanan, S.K., Srinivasan, S., Koti, M.S. et al. A secure hybrid deep learning framework for brain tumor detection and classification. J Big Data 12, 72 (2025). https://doi.org/10.1186/s40537-025-01117-6 [9]Sagar Patil, Mr.Neelesh Tippe, Mr. Pravin Patil, Mr. Chetan Kavade, Mr. Dheeraj Magar, Mr. Nilesh Panari, “Ubiquitous Adoption of Telemedicine to Extend Patient Care beyond the Office,” International Journal of Emerging Engineering Research and Technology, vol. 3(2), pp. 25–28, 2015. [10] S B Patil and R B Kulkarni and S S Patil and P AKharade, “Precision Agriculture Model for Farm Irrigation using Machine Learning to Optimize Water Usage,” IOP Publishing: Earth and Environmental Science, vol. 1285, 2024, doi: 10.1088/1755-1315/1285/1/012017.