Deep Learning-Based Brain Tumor Classification in MRI with External Validation Across Multiple Datasets

Filipe Ferreira Viana; Leandro Henrique Furtado Pinto Silva; João Fernando Mari

doi:10.22456/2175-2745.150908

Authors

Filipe Ferreira Viana Universidade Federal de Viçosa (UFV) https://orcid.org/0009-0007-0052-5957
Leandro Henrique Furtado Pinto Silva Universidade Federal de Viçosa (UFV) https://orcid.org/0000-0002-5765-5206
João Fernando Mari Universidade Federal de Viçosa (UFV) https://orcid.org/0000-0001-8271-3802

DOI:

https://doi.org/10.22456/2175-2745.150908

Keywords:

brain tumor classification, magnetic resonance imaging (MRI), deep learning, data augmentation, external validation

Abstract

The automatic classification of brain tumors in magnetic resonance imaging (MRI) remains a challenging task due to the variability of tumor morphology and the heterogeneity of available datasets. This work presents a comparative evaluation of three deep learning models (ResNet-50, DenseNet-121, and ViT-B16) using the recently curated BRISC 2025 dataset and the BCD-MRI dataset. A systematic evaluation protocol was adopted, including internal validation, external evaluation restricted to test splits, and external evaluation over the entire external datasets. Four levels of data augmentation, ranging from none to strong, were applied to assess their impact on generalization. Internal evaluations achieved high and stable results across all models, while external test set evaluations showed only minor drops. More pronounced differences emerged when models were tested on entire external datasets: knowledge transferred from BCD-MRI generalized better to BRISC than the reverse. Data augmentation reduced part of the performance loss, particularly in external evaluations, yielding significant gains in robustness. Overall, DenseNet-121 proved to be the most reliable architecture, ResNet-50 showed resilience when combined with augmentation, and ViT-B16 achieved strong results in restricted cases but was less stable in whole-dataset external evaluations. The main contribution of this study is to demonstrate that progressive data augmentation enhances the external dataset generalization, highlighting the importance of evaluating models beyond internal splits or limited external test sets to better approximate real-world deployment conditions.

Downloads

Download data is not yet available.

References

[1] ESTEVA, A. et al. A guide to deep learning in healthcare. Nature Medicine, Nature Publishing Group, v. 25, n. 1, p. 24–29, 2019.

[2] LITJENS, G. et al. A survey on deep learning in medical image analysis. Medical Image Analysis, Elsevier, v. 42, p. 60–88, 2017.

[3] GORDILLO, N.; MONTSENY, E.; SOBREVILLA, P. State of the art survey on MRI brain tumor segmentation. Magnetic Resonance Imaging, Elsevier, v. 31, n. 8, p. 1426–1438, 2013.

[4] RONNEBERGER, O.; FISCHER, P.; BROX, T. U-Net: Convolutional networks for biomedical image segmentation. In: SPRINGER. International Conference on Medical Image Computing and Computer-Assisted Intervention. [S.l.], 2015. p. 234–241.

[5] SHORTEN, C.; KHOSHGOFTAAR, T. M. A survey on image data augmentation for deep learning. Journal of Big Data, Springer, v. 6, n. 1, p. 1–48, 2019.

[6] DONG, S.; WANG, P.; ABBAS, K. A survey on deep learning and its applications. Computer Science Review, Elsevier, v. 40, p. 100379, 2021.

[7] SEYREK, E. C.; UYSAL, M. A comparative analysis of various activation functions and optimizers in a convolutional neural network for hyperspectral image classification. Multimedia Tools and Applications, Springer, v. 83, n. 18, p. 53785–53816, 2024.

[8] SILVA, L. H. F. P. et al. Evaluating combinations of optimizers and loss functions for cloud removal using diffusion models. In: Proceedings of the 20th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 3: VISAPP. [s.n.], 2025. p. 648–656. ISBN 978-989-758-728-3. ISSN 2184-4321. Disponível em: https://doi.org/10.5220/0013252100003912.

[9] ZHANG, S. et al. Stroke lesion detection and analysis in MRI images based on deep learning. Journal of Healthcare Engineering, Hindawi, v. 2021, p. 5524769, 2021.

[10] RAHMAN, T.; ISLAM, M. S. MRI brain tumor detection and classification using parallel deep convolutional neural networks. Measurement: Sensors, Elsevier, v. 26, p. 100694, 2023.

[11] ANANTHARAJAN, S. et al. MRI brain tumor detection using deep learning and machine learning approaches. Measurement: Sensors, Elsevier, v. 31, p. 101026, 2024.

[12] ALI, M. A. et al. Enhancing MRI brain tumor classification: A comprehensive approach integrating real-life scenario simulation and augmentation techniques. Physica Medica, Elsevier, v. 127, p. 104841, 2024.

[13] FATEH, A. et al. BRISC: Annotated Dataset for Brain Tumor Segmentation and Classification with Swin-HAFNet. 2025. Disponível em: https://arxiv.org/abs/2506.14318.

[14] KHAN, S. U. R. et al. ShallowMRI: a novel lightweight CNN with novel attention mechanism for multi brain tumor classification in MRI images. Biomedical Signal Processing and Control, Elsevier, v. 111, p. 108425, 2026.

[15] SAFWAN, M. N. et al. T3SSLNet: Triple-method self-supervised learning for enhanced brain tumor classification in MRI. IEEE Access, IEEE, 2025.

[16] KAPLAN. Open Source Dataset, Brain Cancer Detection - MRI Images Dataset. [S.l.]: Roboflow, 2023. https://universe.roboflow.com/kaplan/brain-cancer-detection-mri-images. Visited on 2025-09-01.

[17] HE, K. et al. Deep residual learning for image recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). [S.l.: s.n.], 2016. p. 770–778.

[18] HUANG, G. et al. Densely connected convolutional networks. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). [S.l.: s.n.], 2017. p. 4700–4708.

[19] DOSOVITSKIY, A. et al. An image is worth 16x16 words: Transformers for image recognition at scale. In: International Conference on Learning Representations (ICLR). [S.l.: s.n.], 2021.

[20] VASWANI, A. et al. Attention is all you need. Advances in Neural Information Processing Systems, v. 30, 2017.