Deep Learning-Based Multiclass Classification of Brain Tumors from MRI Images: A CNN Approach

• Author(s): Harshita Rajoria; Sneha Kachhap; Vinisha; Priya
• Paper ID: 1711921
• Page: 574-583
• Published Date: 11-11-2025
• Published In: Iconic Research And Engineering Journals
• Publisher: IRE Journals
• e-ISSN: 2456-8880
• Volume/Issue: Volume 9 Issue 5 November-2025

Abstract

Brain tumor classification using MRI scans is important for early diagnosis and clinical management. This study presents a deep learning-based approach using convolutional neural networks (CNNs) to automatically classify brain MRI images into four distinct categories: glioma, meningioma, pituitary tumor, and no tumor. The proposed methodology employs both custom CNN architectures and transfer learning to enhance performance. Experiments were conducted on a curated dataset comprising 3,500 labeled MRI images, ensuring balanced representation across all classes. Model performance was evaluated using standard metrics including accuracy, precision, recall, and F1-score. The results demonstrate an overall classification accuracy exceeding 90%, with robust performance across all tumor classes. This work illustrates the effectiveness of deep learning for multiclass brain tumor classification and underscores its potential as an efficient, reliable tool for aiding clinical decision-making.

Keywords

Brain Tumor Classification, MRI, Deep Learning, Convolutional Neural Networks, Multiclass Prediction, Transfer Learning, Medical Imaging, Glioma, Meningioma, Pituitary, Automatic Diagnosis

Citations

IRE Journals:
Harshita Rajoria, Sneha Kachhap, Vinisha, Priya "Deep Learning-Based Multiclass Classification of Brain Tumors from MRI Images: A CNN Approach" Iconic Research And Engineering Journals Volume 9 Issue 5 2025 Page 574-583 https://doi.org/10.64388/IREV9I5-1711921

IEEE:
Harshita Rajoria, Sneha Kachhap, Vinisha, Priya "Deep Learning-Based Multiclass Classification of Brain Tumors from MRI Images: A CNN Approach" Iconic Research And Engineering Journals, 9(5) https://doi.org/10.64388/IREV9I5-1711921