Ensembling Transformer-Based Models for 3D Ischemic Stroke Segmentation in Non-Contrast CT

Ensembling Transformer-Based Models for 3D Ischemic Stroke Segmentation in Non-Contrast CT Full article

Journal

Applied Sciences
, E-ISSN: 2076-3417

Output data

Year: 2025, Volume: 15, Number: 17, Article number : 9725, Pages count : 20 DOI: 10.3390/app15179725

Tags

UNETR; Swin Transformer; CT; ischemic stroke; deep learning; segmentation; ensemble of models

Authors

Cherikbayeva Lyailya ¹ , Berikov Vladimir ^2,3 , Melis Zarina ¹ , Yeleussinov Arman ¹ , Baigozhanova Dametken ⁴ , Tasbolatuly Nurbolat ^5,4 , Temirbekova Zhanerke ¹ , Mikhailapov Denis ^2,3

Affiliations

1	Department of Computer Science, Al-Farabi Kazakh National University, Almaty 050040, Kazakhstan
2	Department of Mechanics and Mathematics, Novosibirsk State University, 630090 Novosibirsk, Russia
3	Sobolev Institute of Mathematics SB RAS, 630090 Novosibirsk, Russia
4	Higher School of Information Technology and Engineering, Astana International University, Astana 010000, Kazakhstan
5	Department of Computer Engineering, Astana IT University, Astana 010000, Kazakhstan

Ischemic stroke remains one of the leading causes of mortality and disability, and accurate segmentation of the affected areas on CT brain images plays a crucial role in timely diagnosis and clinical decision-making. This study proposes an ensemble approach based on the combination of the transformer-based models SE-UNETR and Swin UNETR using a weighted voting strategy. Its performance was evaluated using the Dice similarity coefficient, which quantifies the overlap between the predicted lesion regions and the ground-truth annotations. In this study, three-dimensional CT scans of the brain from 98 patients with a confirmed diagnosis of acute ischemic stroke were used. The data were provided by the International Tomography Center, SB RAS. The experimental results demonstrated that the ensemble based on transformer models significantly outperforms each individual model, providing more stable and accurate predictions. The final Dice coefficient reached 0.7983, indicating the high effectiveness of the proposed approach for ischemic lesion segmentation in CT images. The analysis showed more precise delineation of ischemic lesion boundaries and a reduction in segmentation errors. The proposed method can serve as an effective tool in automated stroke diagnosis systems and other applications requiring high-accuracy medical image analysis.

Cherikbayeva L. , Berikov V. , Melis Z. , Yeleussinov A. , Baigozhanova D. , Tasbolatuly N. , Temirbekova Z. , Mikhailapov D.
Ensembling Transformer-Based Models for 3D Ischemic Stroke Segmentation in Non-Contrast CT
Applied Sciences. 2025. V.15. N17. 9725 :1-20. DOI: 10.3390/app15179725 WOS Scopus OpenAlex

Submitted:	Jul 22, 2025
Accepted:	Aug 27, 2025
Published print:	Sep 4, 2025
Published online:	Sep 4, 2025

Web of science:	WOS:001569550400001
Scopus:	2-s2.0-105015582830
OpenAlex:	W4413989486

Пока нет цитирований