A multimodal prediction model for brain hematoma expansion based on deep neural network with transformer-driven collaborative attention fusion
Medical & Biological Engineering & Computing · 2026
DOI: 10.1007/s11517-026-03643-8 ↗Kurz gefasst
Fachartikel in der Zeitschrift Medical & Biological Engineering & Computing, 2026. Verfasst von Nanxu Tao und 8 weiteren Autor:innen. Beteiligte Einrichtung: School of Electronic and Information Engineering, Suzhou University of Science and Technology, Suzhou, Jiangsu, 215009, China. Hauptthemen laut MEDLINE: Hematoma, Cerebral Hemorrhage, Deep Learning, Neural Networks, Computer. Die Arbeit nennt 38 Literaturangaben. Gefördert von: National Natural Science Foundation of China.
Abstract (Originalsprache)
Hematoma expansion (HE) in patients with spontaneous intracerebral hemorrhage (ICH) is strongly associated with death and disability, so accurate early prediction is important for clinical decision making. Existing methods often rely on single imaging signs or traditional scoring systems and are unable to consider both the three dimensional hematoma structure and the clinical status at the same time, which limits predictive performance and robustness. To address this, we propose a multimodal deep learning model named SResFormer based on position-aware Computed Tomography (CT) feature encoding. CT images and parenchymal hematoma masks are processed in a dual channel scheme. Spatial order across slices is then modeled using position-aware CT feature encoding, enabling explicit capture of three dimensional patterns of hematoma evolution. These representations are subsequently fused with structured clinical features through semantic alignment to achieve integrated cross modality prediction. SResFormer demonstrates significantly superior performance compared with traditional scoring systems, unimodal Convolutional Neural Network (CNN) based methods, and shallow multimodal fusion approaches on multicenter datasets. Ablation studies confirm both the independent contributions and the synergistic benefits of each module. SHapley Additive exPlanations (SHAP) analysis further indicates that the model reasoning is consistent with established medical knowledge.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- School of Electronic and Information Engineering, Suzhou University of Science and Technology, Suzhou, Jiangsu, 215009, China.
- Chongqing Industrial Big Data Innovation Center Co., Ltd., Chongqing, 400707, China.
- Department of Neurosurgery, The Second Affiliated Hospital of Soochow University, Suzhou, 215004, China.
- Department of Neurosurgery, The Second Affiliated Hospital of Soochow University, Suzhou, 215004, China. chenyanming1@126.com.
- School of Electronic and Information Engineering, Suzhou University of Science and Technology, Suzhou, Jiangsu, 215009, China. fqm_1@126.com.
Förderung
- National Natural Science Foundation of China
Zitieren
APA
Tao, N., Chen, J., Wu, H., Dai, J., Zhong, T., Song, C., Wang, Y., Chen, Y., & Fu, Q. (2026). A multimodal prediction model for brain hematoma expansion based on deep neural network with transformer-driven collaborative attention fusion. Medical & Biological Engineering & Computing. https://doi.org/10.1007/s11517-026-03643-8
Vancouver
Tao N, Chen J, Wu H, Dai J, Zhong T, Song C, et al. A multimodal prediction model for brain hematoma expansion based on deep neural network with transformer-driven collaborative attention fusion. Medical & Biological Engineering & Computing. 2026. doi:10.1007/s11517-026-03643-8
BibTeX
@article{Tao2026,
author = {Tao, Nanxu and Chen, Jianping and Wu, Hongjie and Dai, Jiaqi and Zhong, Tao and Song, Chenyang and Wang, Ye and Chen, Yanming and Fu, Qiming},
title = {A multimodal prediction model for brain hematoma expansion based on deep neural network with transformer-driven collaborative attention fusion},
journal = {Medical & Biological Engineering & Computing},
year = {2026},
doi = {10.1007/s11517-026-03643-8}
}
EUVIMED