Expert-Guided Cross-View Fusion With Self-Derived Lesion Proposals for Multi-View Diabetic Retinopathy Grading
IEEE Transactions on Image Processing · 2026 · 35 · 6446-6459
DOI: 10.1109/tip.2026.3699089 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Image Processing, 2026. Verfasst von Wai Keung Wong und 5 weiteren Autor:innen. Hauptthemen laut MEDLINE: Diabetic Retinopathy, Image Interpretation, Computer-Assisted. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 62 Literaturangaben. Gefördert von: Laboratory for Artificial Intelligence in Design, the InnoHK Initiative of the Innovation and Technology Commission of Hong Kong Special Administrative Region…, Basic and Applied Basic Research Foundation of Guangdong Province, Open Project Program of State Key Laboratory of Virtual Reality Technology and Systems, Beihang University,.
Abstract (Originalsprache)
Recent advances in multi-view fundus imaging show great promise for automated diabetic retinopathy (DR) grading. However, mainstream end-to-end CNN/Transformer pipelines rely on striding or tokenization that compresses spatial detail, causing small, low-contrast lesions (e.g., microaneurysms) to be under-represented and creating performance ceilings. Prior efforts have mitigated this by incorporating external lesion- or vessel-level annotations into models. However, such labels are costly to acquire, break the end-to-end training, and make performance over-reliant on the annotation quality. To reduce dependence on expensive annotations, we propose an end-to-end framework that generates lesion proposals on the fly during training and inference, providing self-derived cues for grading. First, we introduce a Grade-Activated Lesion Proposal (GALP) module that derives grade-conditioned evidence maps (GEMs) from stage-wise auxiliary classifiers and selects the top-K high-evidence regions per view as lesion proposals. Second, we propose a Cross-View Lesion Expert Guided Regional Fusion (LGRF) module, which selectively activates experts for a view's lesion proposals based on contextual guidance from other views, ensuring that only the most relevant feature extractors contribute to fusion. Experimental results on two multi-view DR datasets show that our method matches or surpasses strong baselines without external annotations, confirming that self-generated proposals can substantially reduce annotation needs.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Förderung
- Laboratory for Artificial Intelligence in Design, the InnoHK Initiative of the Innovation and Technology Commission of Hong Kong Special Administrative Region…
- Basic and Applied Basic Research Foundation of Guangdong Province
- Open Project Program of State Key Laboratory of Virtual Reality Technology and Systems, Beihang University,
- Ministry of Education’s Industry-University Cooperation and Collaborative Education Project
Zitieren
APA
Wong, W. K., Liu, W., Zhou, X., Hou, J., Lin, Y., & Wen, J. (2026). Expert-Guided Cross-View Fusion With Self-Derived Lesion Proposals for Multi-View Diabetic Retinopathy Grading. IEEE Transactions on Image Processing, 35, 6446-6459. https://doi.org/10.1109/tip.2026.3699089
Vancouver
Wong WK, Liu W, Zhou X, Hou J, Lin Y, Wen J. Expert-Guided Cross-View Fusion With Self-Derived Lesion Proposals for Multi-View Diabetic Retinopathy Grading. IEEE Transactions on Image Processing. 2026;35:6446-6459. doi:10.1109/tip.2026.3699089
BibTeX
@article{Wong2026,
author = {Wong, Wai Keung and Liu, Wenzhe and Zhou, Xueling and Hou, Junlin and Lin, Yuxin and Wen, Jie},
title = {Expert-Guided Cross-View Fusion With Self-Derived Lesion Proposals for Multi-View Diabetic Retinopathy Grading},
journal = {IEEE Transactions on Image Processing},
year = {2026},
volume = {35},
pages = {6446--6459},
doi = {10.1109/tip.2026.3699089}
}
EUVIMED