Brain Tumor Classification with Hybrid Deep Learning Models from MRI Images

Dhouha Boubdellaha, Raouia Mokni, Boudour Ammar, Boudour Ammar

2025

Abstract

Brain tumor classification using MRI images plays a crucial role in medical diagnostics, enabling early detection and improving treatment planning. Traditional diagnostic methods are often subjective and time-intensive, emphasizing the need for automated and precise solutions. This study explores hybrid deep learning models alongside fine-tuned pre-trained architectures for classifying brain tumors into four categories: glioma, meningioma, pituitary tumor, and healthy brain tissue. The proposed approach incorporates both hybrid ensemble models—Ens-VGG16-FT-InceptionV3, Ens-ViT-FT-InceptionV3, and Ens-CNN-ViT—and fine-tuned architectures—FT-VGG16, FT-VGG19, and FT-InceptionV3. To enhance model robustness and generalization, data augmentation techniques such as rotation and scaling were applied. Among these models, the hybrid ensemble Ens-VGG16-FT-InceptionV3 achieved the highest accuracy and F1-score of 99%, outperforming both standalone models and other hybrid configurations. These findings demonstrate the effectiveness of integrating complementary architectures for improved brain tumor classification. Ultimately, this study highlights the potential of hybrid ensemble learning to advance brain tumor diagnostics, providing more accurate, reliable, and scalable medical imaging solutions.

Paper Citation

in Harvard Style

Boubdellaha D., Mokni R. and Ammar B. (2025). Brain Tumor Classification with Hybrid Deep Learning Models from MRI Images. In Proceedings of the 14th International Conference on Data Science, Technology and Applications - Volume 1: DATA; ISBN 978-989-758-758-0, SciTePress, pages 455-464. DOI: 10.5220/0013523700003967

in Bibtex Style

@conference{data25,
author={Dhouha Boubdellaha and Raouia Mokni and Boudour Ammar},
title={Brain Tumor Classification with Hybrid Deep Learning Models from MRI Images},
booktitle={Proceedings of the 14th International Conference on Data Science, Technology and Applications - Volume 1: DATA},
year={2025},
pages={455-464},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013523700003967},
isbn={978-989-758-758-0},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 14th International Conference on Data Science, Technology and Applications - Volume 1: DATA
TI - Brain Tumor Classification with Hybrid Deep Learning Models from MRI Images
SN - 978-989-758-758-0
AU - Boubdellaha D.
AU - Mokni R.
AU - Ammar B.
PY - 2025
SP - 455
EP - 464
DO - 10.5220/0013523700003967
PB - SciTePress