Four-Class Brain Tumor Classification Using ConvNeXt with Grad-CAM-Based Explainable

Ni Kadek Jegeg Anastasya Dwipayanti; I Gusti Ngurah Lanang Wijayakusuma

doi:10.30871/jaic.v10i4.13607

Authors

Ni Kadek Jegeg Anastasya Dwipayanti Universitas Udayana
I Gusti Ngurah Lanang Wijayakusuma Universitas Udayana

DOI:

https://doi.org/10.30871/jaic.v10i4.13607

Keywords:

CNN, MRI Classification, ConvNeXt, Brain Tumor, Grad-CAM

Abstract

Brain tumor is one of the most critical diseases with high mortality rates requiring early and accurate diagnosis. This study proposes a deep learning-based brain tumor classification system using the ConvNeXt-Base architecture to classify four categories: glioma, meningioma, pituitary tumor, and no tumor. To further advance this field, this study addresses two open challenges in prior literature, the need for larger and more diverse datasets to ensure model robustness, and the demand for model transparency (resolving the black-box problem) to facilitate clinical adoption.. To address these, this study combines MRI images from three public sources, Figshare (Cheng et al., 2016), Br35H (Hamada, 2020), and Mendeley (Hira et al., 2025), totaling 11,474 unique images after perceptual hash-based deduplication. The model was trained using progressive unfreezing with AdamW optimizer, CosineAnnealingLR scheduler, and class-weighted cross-entropy loss. Gradient-weighted Class Activation Mapping (Grad-CAM) is integrated as an Explainable AI (XAI) approach to visualize the regions influencing model decisions. The proposed ConvNeXt-Base model achieved an accuracy of 99.25%, macro precision of 99.23%, macro recall of 99.22%, macro F1-score of 99.22%, and macro AUC-ROC of 0.9995 on the test set. Grad-CAM visualizations confirm that the model focuses on clinically relevant tumor regions, particularly the sella turcica for pituitary tumors and distinct mass boundaries for meningioma, thereby improving interpretability and clinical trustworthiness of the AI system.

Downloads

Download data is not yet available.

References

[1] V. Essianda, A. D. Indrasari, P. Widyastuti, T. Syahla, and R. Rohadi, “Brain Tumor : Molecular Biology, Pathophysiology, and Clinical Symptoms,” J. Biol. Trop., vol. 23, no. 4, pp. 260–269, 2023, doi: 10.29303/jbt.v23i4.5585.

[2] R. N. Asif et al., “Brain tumor detection empowered with ensemble deep learning approaches from MRI scan images,” Sci. Rep., vol. 15, no. 1, pp. 1–22, 2025, doi: 10.1038/s41598-025-99576-7.

[3] D. Rastogi et al., “Dual-stage deep learning framework for brain tumor classification and localization using multimodal MRI scans,” Intell. Med., vol. 13, no. February, 2026, doi: 10.1016/j.ibmed.2026.100361.

[4] Y. Wong, E. L. M. Su, C. F. Yeong, W. Holderbaum, and C. Yang, “Brain tumor classification using MRI images and deep learning techniques,” PLoS One, vol. 20, no. 5 May, pp. 1–19, 2025, doi: 10.1371/journal.pone.0322624.

[5] S. A. Shreya, M. A. I. Siddique, and S. Tasnim, “DB-FGA-Net: Dual Backbone Frequency Gated Attention Network for Multi-Class Brain Tumor Classification with Grad-CAM Interpretability,” pp. 1–25, 2025, [Online]. Available: https://arxiv.org/pdf/2510.20299

[6] K. K. A. N. D. K. Jyoti, “Recent Advancements in Grad-CAM and variants: Enhancing Brain Tumor Detection, Segmentation, and Classification,” Res. Sq. (Research Square), pp. 1–18, 2024, [Online]. Available: https://doi.org/10.21203/rs.3.rs-5485128/v1

[7] O. P. Mmileng, A. Whata, M. Olusanya, and S. Mhlongo, “Application of ConvNeXt with transfer learning and data augmentation for malaria parasite detection in resource-limited settings using microscopic images,” PLoS One, vol. 20, no. 6 June, pp. 1–39, 2025, doi: 10.1371/journal.pone.0313734.

[8] G. Ekayanda and M. Rahardi, “Analysis of Deep Learning Algorithms Using ConvNeXt and Vision Transformer for Brain Tumor Disease,” J. Appl. Informatics Comput., vol. 9, no. 6, pp. 3295–3304, 2025, doi: 10.30871/jaic.v9i6.11438.

[9] K. Pant, P. K. Dutta Pramanik, and Z. Zhao, “A Robust ConvNeXt-Based Framework for Efficient, Generalizable, and Explainable Brain Tumor Classification on MRI,” Bioengineering, vol. 13, no. 2, 2026, doi: 10.3390/bioengineering13020157.

[10] J. Cheng, “Brain Tumor Dataset,” 2024, Figshare. doi: https://doi.org/10.6084/m9.figshare.1512427.

[11] A. Hamada, “Br35H :: Brain Tumor Detection,” 2020, Kaggle.

[12] M. I. K. Hira, M. S. Hossain, and M. M. A. Bithee, “Brain Tumor MRI Dataset (Glioma, Meningioma, Pituitary, No Tumor),” 2025, Mendeley Data. doi: 10.17632/zwr4ntf94j.1.

[13] D. L. S. Dimara, S. W. Putri, R. Amelia, Z. I. Arishandy, and A. M. Rizki, “Penerapan Convolutional Neural Network (CNN) dalam Klasifikasi Citra MRI untuk Deteksi Tumor Otak Manusia,” KERNEL J. Ris. Inov. Bid. Inform. dan Pendidik. Inform., vol. 4, no. 2, pp. 70–77, 2023, doi: 10.31284/j.kernel.2023.v4i2.6960.

[14] Y. Liu, R. T. Collins, W. E. Rothfus, and M. D, “Automatic Extraction of the Central Symmetry (Mid-Sagittal) Plane from Neuroradiology Images,” Carnegie Mellon Univ. Robot. Inst. Tech. Report, C., pp. 1–29, 1996.

[15] E. Goceri, Medical image data augmentation: techniques, comparisons and interpretations, vol. 56, no. 11. Springer Netherlands, 2023. doi: 10.1007/s10462-023-10453-z.

[16] B. S. Abd El-Wahab, M. E. Nasr, S. Khamis, and A. S. Ashour, “BTC-fCNN: Fast Convolution Neural Network for Multi-class Brain Tumor Classification,” Heal. Inf. Sci. Syst., vol. 11, no. 1, pp. 1–22, 2023, doi: 10.1007/s13755-022-00203-w.

[17] R. R. Selvaraju, M. Cogswell, A. Das, R. Vedantam, D. Parikh, and D. Batra, “Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization,” Int. J. Comput. Vis., vol. 128, no. 2, pp. 336–359, 2020, doi: 10.1007/s11263-019-01228-7.

[18] Z. Liu, H. Mao, C. Y. Wu, C. Feichtenhofer, T. Darrell, and S. Xie, “A ConvNet for the 2020s,” Proc. IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognit., vol. 2022-June, pp. 11966–11976, 2022, doi: 10.1109/CVPR52688.2022.01167.