Copyright © 2025 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{183471,
        author = {Umbare Rupali Tukaram},
        title = {CNN and Transfer Learning Techniques for Improved Brain Tumor Classification from MRI},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {12},
        number = {3},
        pages = {1659-1665},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=183471},
        abstract = {Brain tumors are among the most dangerous types of cancer and highlight the importance of a timely and accurate diagnosis. An example of a non-invasive, 1-source modality to identify and assess brain tumors is Magnetic Resonance Imaging (MRI). Interpreting MRI data manually via visual inspection is time-consuming and can be error-prone. To that end, this study proposes an automated classification framework with deep learning models such as Convolutional Neural Networks (CNN), and hybrid approaches of CNN and Support Vector Machine (CNN+SVM) and CNN and the k-nearest neighbors algorithm (CNN+KNN). Transfer learning is also used by using fine-tuned pre-trained networks of InceptionV3 and Xception. The experiments were performed on both the Sartaj dataset and the BraTS dataset. The results indicate that transfer learning can aid my classification performance, with Xception achieving a maximum accuracy of 94.1%. Furthermore, the proposed approach demonstrates statistically reliable and robust results.},
        keywords = {Brain tumor, BraTS, Convolutional Neural Networks (CNN), Deep learning},
        month = {August},
        }