Accurate segmentation of brain tumors is a critical component of image-guided diagnosis and treatment. However, this task faces substantial challenges due to highly irregular morphology, ill-defined boundaries, strong tissue heterogeneity of tumor subregions, as well as distribution discrepancies and missing modalities in multi-modal MRI data. In recent years, Transformer-based methods, driven by global self-attention and flexible cross-modal interaction mechanisms, have achieved remarkable progress in brain tumor segmentation. This paper systematically reviews representative Transformer-based approaches for multi-modal MRI brain tumor segmentation over the past five years, focusing on architectural design, problem–solution correspondence, applicable scenarios, and quantitative metrics. The analysis reveals two primary research trajectories: one centered on improving segmentation accuracy and boundary quality via multi-scale attention, cross-modal fusion, and explicit boundary modeling; the other oriented toward clinical needs, emphasizing interpretable outputs, robustness under missing modalities, and integrated segmentation–diagnosis pipelines. Evaluations on public datasets such as BraTS demonstrate that Transformers excel in global consistency modeling and small-lesion perception, while boundary errors (HD95) and stability under worst-case modality combinations remain major bottlenecks. Future research should prioritize multi-center domain generalization, lightweight deployment, and trustworthy segmentation mechanisms to drive the transition from benchmark leadership to meaningful clinical deployment.

Dai Lina, Zheng Donghua, Xiu Weirong, Ye Lizhu. Research on Transformer-Based Brain Tumor Image Segmentation. Academic Journal of Computing & Information Science (2026), Vol. 9, Issue 2: 23-29. https://doi.org/10.25236/AJCIS.2026.090204.

[1] de Verdier, M.C., et al., The 2024 brain tumor segmentation (brats) challenge: Glioma segmentation on post-treatment mri. arXiv preprint arXiv:2405.18368, 2024.

[2] Wang, P., et al., Vision transformers in multi-modal brain tumor MRI segmentation: A review. Meta-Radiology, 2023. 1(1): p. 100004.

[3] She, D., et al. Eoformer: Edge-oriented transformer for brain tumor segmentation. in International Conference on Medical Image Computing and Computer-Assisted Intervention. 2023. Springer.

[4] Zeineldin, R.A., et al., Explainable hybrid vision transformers and convolutional network for multimodal glioma segmentation in brain MRI. Scientific reports, 2024. 14(1): p. 3713.

[5] Guo, C., et al., AMGFormer: Adaptive Multi-Granular Transformer for Brain Tumor Segmentation with Missing Modalities. arXiv preprint arXiv:2601.19349, 2026.

[6] Kumar, A., A comprehensive review of transformer models in brain tumor analysis. Expert Systems with Applications, 2025: p. 130509.

[7] Bonato, B., L. Nanni, and A. Bertoldo, Advancing precision: A comprehensive review of MRI segmentation datasets from brats challenges (2012–2025). Sensors (Basel, Switzerland), 2025. 25(6): p. 1838.

[8] Mehta, R., et al., QU-BraTS: MICCAI BraTS 2020 challenge on quantifying uncertainty in brain tumor segmentation-analysis of ranking scores and benchmarking results. The journal of machine learning for biomedical imaging, 2022. 2022: p. https://www. melba-journal. org/papers/2022: 026. html.

[9] Wang, W., et al. Transbts: Multimodal brain tumor segmentation using transformer. in International conference on medical image computing and computer-assisted intervention. 2021. Springer.

[10] Hatamizadeh, A., et al. Swin unetr: Swin transformers for semantic segmentation of brain tumors in mri images. in International MICCAI brainlesion workshop. 2021. Springer.

[11] Yang, F., et al., HCA-former: Hybrid convolution attention transformer for 3D medical image segmentation. Biomedical Signal Processing and Control, 2024. 90: p. 105834.

[12] Liu, T., et al., VSmTrans: A hybrid paradigm integrating self-attention and convolution for 3D medical image segmentation. Medical image analysis, 2024. 98: p. 103295.

[13] Jagadeesh, B. and G.A. Kumar, Brain tumor segmentation with missing MRI modalities using edge aware discriminative feature fusion based transformer U-net. Applied Soft Computing, 2024. 161: p. 111709.

[14] Zhu, H., et al., multiPI-TransBTS: A multi-path learning framework for brain tumor image segmentation based on multi-physical information. Computers in Biology and Medicine, 2025. 191: p. 110148.

[15] Lv, C., et al., BrainTumNet: multi-task deep learning framework for brain tumor segmentation and classification using adaptive masked transformers. Frontiers in Oncology, 2025. 15: p. 1585891.

[16] Huang, Y., L. Chen, and C. Zhou, Multi-Modal Brain Tumor Segmentation via 3D Multi-Scale Self-attention and Cross-attention. arXiv preprint arXiv:2504.09088, 2025.