Citation

Shahriar, Muhammad Muhtasim and Golam Hafiz, M. M. and Rabbi, Riadul Islam and Liew, Tze Wei and Tarin, Fatema Mostafa and Haque Tusher, Ekramul (2025) Comparative Analysis of Deep Learning Architectures for Brain Tumor Classification using MRI Images. In: 8th 2025 International Conference on New Media Studies, CONMEDIA 2025, 14 October 2025 - 17 October 2025, Malacca, Malaysia.

Text IEEE Xplore Full-Text PDF_21.pdf - Published Version Restricted to Repository staff only Download (1MB)

Abstract

Accurate detection and classification of brain tumors by Magnetic Resonance Imaging is crucial for early diagnosis and treatment planning but is still challenging due to variability in tumor heterogeneity and diagnostic subjectivity. This work offers a comparative analysis of Convolutional Neural Network-based architectures such as VGG16/19, ResNet50/152, DenseNet121/201, InceptionV3, EfficientNet-B4/B7, and a Custom CNN-7. Using a Kaggle MRI dataset across four tumor classes (glioma, meningioma, pituitary, and no tumor), DenseNet121 and InceptionV3 achieved the best results with 99.85% accuracy. DenseNet201 (99.31%) and ResNet152 (99.08%) closely followed. These models consistently demonstrated ≈ 0.99 precision, recall, and F1-score, outperforming traditional CNN baselines. This study highlights the balance between accuracy and computational efficiency across multiple architectures, underscoring the potential of CNNs and transfer learning for clinical applications in brain tumor diagnosis.

Item Type:	Conference or Workshop Item (Paper)
Uncontrolled Keywords:	Brain tumor, deep learning
Subjects:	R Medicine > R Medicine (General) > R858-859.7 Computer applications to medicine. Medical informatics
Divisions:	Faculty of Information Science and Technology (FIST)
Depositing User:	Ms Rosnani Abd Wahab
Date Deposited:	20 Apr 2026 04:29
Last Modified:	20 Apr 2026 04:29
URII:	http://shdl.mmu.edu.my/id/eprint/15788

Downloads

Edit (login required)