Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{206607,
        author = {VARAPRASAD K S B and Mr. PRASADA RAO PEERI},
        title = {Deep Neural Framework for Automated Brain Tumor Identification in MRI Scans},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {13},
        number = {2},
        pages = {2249-2258},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=206607},
        abstract = {Accurate identification of brain tumors from magnetic resonance imaging (MRI) scans plays a significant role in supporting clinical diagnosis and treatment planning. Manual examination of large volumes of MRI images is time-consuming and may be influenced by inter-observer variability, creating a need for reliable computer-aided diagnostic systems. This study presents a deep learning-based framework for automated brain tumor classification using MRI images. The proposed work investigates the performance of a custom Convolutional Neural Network (CNN) and three transfer learning models, namely ResNet18, VGG16, and ResNet50, for the classification of four categories: glioma, meningioma, pituitary tumor, and no tumor. Image preprocessing techniques, including resizing, normalization, and data augmentation, were applied to improve model robustness and generalization. To enhance transparency in decision-making, Gradient-weighted Class Activation Mapping (Grad-CAM) was incorporated to visualize the image regions that contributed most to the model predictions. Experimental evaluation demonstrated that ResNet18 achieved superior performance compared to the other models, obtaining a test accuracy of 94.39%. The proposed framework not only provides accurate classification results but also offers visual interpretability, making it more suitable for medical imaging applications. Furthermore, a user-friendly Gradio-based interface was developed to facilitate real-time prediction of brain tumor categories from MRI scans. The findings indicate that combining transfer learning with explainable artificial intelligence can contribute to the development of effective and trustworthy computer-aided diagnostic systems for brain tumor analysis.},
        keywords = {Brain Tumor Classification, Magnetic Resonance Imaging (MRI), Deep Learning, Convolutional Neural Network (CNN), Transfer Learning, ResNet18, VGG16, ResNet50, Grad-CAM, Explainable Artificial Intelligence (XAI), Medical Image Analysis.},
        month = {July},
        }