Application of Vision Transformer for Brain Stroke Classification Based on CT Images
Received: 30 June 2025 | Revised: 18 July 2025 and 27 August 2025 | Accepted: 2 September 2025 | Online: 16 October 2025
Corresponding author: Azhar Tursynova
Abstract
The development of artificial intelligence and machine learning has had a significant impact on medical diagnostics. This paper examines the application of the Vision Transformer (ViT) architecture for the task of classifying Computed Tomography (CT) images of the brain for the presence or absence of stroke signs. ViT which was originally developed for computer vision tasks, uses attention mechanisms, allowing the model to focus on important aspects of an image without first learning using task-specific data. The article presents the results of experiments on training the ViT model on a dataset of CT images, as well as performance comparisons with traditional methods such as Convolutional Neural Networks (CNNs). The results show that ViT can effectively classify strokes, demonstrating high accuracy and generalization ability based on new data. These findings may contribute to the wider application of transformer models in medical imaging which in the future may improve diagnostic accuracy and accelerate the treatment of stroke patients.
Keywords:
brain stroke, vision transformer, CNN, CT, classification, DLDownloads
References
P. Vilela and H. A. Rowley, "Brain ischemia: CT and MRI techniques in acute ischemic stroke," European Journal of Radiology, vol. 96, pp. 162–172, Nov. 2017. DOI: https://doi.org/10.1016/j.ejrad.2017.08.014
L. Alzubaidi et al., "Review of deep learning: concepts, CNN architectures, challenges, applications, future directions," Journal of Big Data, vol. 8, no. 1, Mar. 2021, Art. no. 53. DOI: https://doi.org/10.1186/s40537-021-00444-8
Y. Liu et al., "A Survey of Visual Transformers," IEEE Transactions on Neural Networks and Learning Systems, vol. 35, no. 6, pp. 7478–7498, Jun. 2024. DOI: https://doi.org/10.1109/TNNLS.2022.3227717
C.-L. Chin et al., "An automated early ischemic stroke detection system using CNN deep learning algorithm," in 2017 IEEE 8th International Conference on Awareness Science and Technology (iCAST), Taichung, Taiwan, Aug. 2017, pp. 368–372. DOI: https://doi.org/10.1109/ICAwST.2017.8256481
A. Dosovitskiy et al., "An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale." arXiv, Jun. 03, 2021.
R. Yacouby and D. Axman, "Probabilistic Extension of Precision, Recall, and F1 Score for More Thorough Evaluation of Classification Models," in Proceedings of the First Workshop on Evaluation and Comparison of NLP Systems, Online, Aug. 2020, pp. 79–91. DOI: https://doi.org/10.18653/v1/2020.eval4nlp-1.9
A. Rahman, "Brain Stroke CT Image Dataset." 2021, [Online]. Available: https://www.kaggle.com/datasets/afridirahman/brain-stroke-ct-image-dataset.
A. Tursynova et al., "Deep Learning-Enabled Brain Stroke Classification on Computed Tomography Images," Computers, Materials and Continua, vol. 75, no. 1, pp. 1431–1446, Jan. 2023. DOI: https://doi.org/10.32604/cmc.2023.034400
T. Rohini and P. Praveen, "An Intuitive Approach on Transfer Learning with an IBF+IHP Model for Stroke Classification and Prediction," Engineering, Technology & Applied Science Research, vol. 15, no. 1, pp. 19655–19660, Feb. 2025. DOI: https://doi.org/10.48084/etasr.9031
M. H. Ferris et al., "Using ROC curves and AUC to evaluate performance of no-reference image fusion metrics," in 2015 National Aerospace and Electronics Conference (NAECON), Dayton, OH, USA, Jun. 2015, pp. 27–34. DOI: https://doi.org/10.1109/NAECON.2015.7443034
J. Miao and W. Zhu, "Precision–recall curve (PRC) classification trees," Evolutionary Intelligence, vol. 15, no. 3, pp. 1545–1569, Sep. 2022. DOI: https://doi.org/10.1007/s12065-021-00565-2
Downloads
How to Cite
License
Copyright (c) 2025 Azhar Tursynova
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain the copyright and grant the journal the right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) after its publication in ETASR with an acknowledgement of its initial publication in this journal.
