Problem
Research questions and friction points this paper is trying to address.
Stroke Detection
Retinal Fundus Imaging
Transient Ischemic Attack
Multi-view Imaging
Braided Vision Transformer for Stroke Detection in Multi-view Retinal Fundus Imaging
🤖 AI Summary
This study addresses the need for non-invasive screening of stroke and transient ischemic attack by proposing the Braided Vision Transformer (BViT). As the first retinal Vision Transformer applied to stroke assessment, BViT innovatively integrates binocular multi-view fundus image features through cross-view interaction modeling to precisely extract cerebrovascular biomarkers. Evaluated on a proprietary Stroke-Data dataset, the model achieved an AUC of 0.75, demonstrating performance significantly superior to conventional Vision Transformers. These results effectively validate both the technical advantages and clinical potential of multi-view fundus imaging for rapid stroke detection, offering a promising non-invasive approach for early cerebrovascular risk assessment.
📝 Abstract
Stroke remains a leading cause of mortality and morbidity worldwide, emphasizing the importance of its accurate and immediate assessment. Retinal fundus imaging has emerged as a promising modality for stroke assessment, as the retina reflects cerebrovascular and neurological risk factors. Contrary to conventional neuroimaging techniques, retinal fundus imaging offers a non-invasive, cost-effective, and portable alternative for rapid screening. This paper explores the feasibility of retinal fundus imaging for stroke and transient ischemic attack (TIA) detection using macula-centric and optic nerve head-centric views captured from both eyes. Our study introduces, to the best of our knowledge, the first vision transformer model for retinal fundus imaging in stroke assessment, offering a novel approach for capturing retinal patterns. Thereby, we propose the Braided Vision Transformer (BViT) model, which extracts representative features from the given multi-view images while simultaneously capturing inter-view relationships across both eyes, enabling a more informative understanding of retinal biomarkers associated with cerebrovascular events. Experiments conducted on our collected Stroke-Data dataset demonstrate that BViT achieves an AUC score of 0.75 for stroke detection, outperforming regular vision transformers.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Braided Vision Transformer
Multi-view Retinal Fundus Imaging
Stroke Detection
Inter-view Relationships
Retinal Biomarkers
🔎 Similar Papers
No similar papers found.
