Problem
Research questions and friction points this paper is trying to address.
Primary Open-Angle Glaucoma
Severity Assessment
Fundus Photography
Explainable Severity ranking via pairwise n-hidden comparison: a case study of glaucoma
🤖 AI Summary
To address the challenge of quantitatively assessing disease severity during the asymptomatic progression stage of primary open-angle glaucoma (POAG), this paper proposes an interpretable severity ranking framework based on fundus photographs. Methodologically, we introduce a novel Siamese network–driven pairwise latent-space comparison mechanism to enable fine-grained inter-patient severity ranking; further, we integrate multi-scale feature modeling with Grad-CAM–enhanced saliency explanation to generate attribution heatmaps aligned with clinical reasoning. Our key contributions are threefold: (1) the first deep integration of pairwise learning and model interpretability, jointly optimizing ranking accuracy and clinical trustworthiness; (2) statistically significant improvements in ranking accuracy over conventional models on POAG severity assessment; and (3) high spatial concordance between generated saliency maps and expert annotations—demonstrating enhanced physician trust and decision-support utility.
📝 Abstract
Primary open-angle glaucoma (POAG) is a chronic and progressive optic nerve condition that results in an acquired loss of optic nerve fibers and potential blindness. The gradual onset of glaucoma results in patients progressively losing their vision without being consciously aware of the changes. To diagnose POAG and determine its severity, patients must undergo a comprehensive dilated eye examination. In this work, we build a framework to rank, compare, and interpret the severity of glaucoma using fundus images. We introduce a siamese-based severity ranking using pairwise n-hidden comparisons. We additionally have a novel approach to explaining why a specific image is deemed more severe than others. Our findings indicate that the proposed severity ranking model surpasses traditional ones in terms of diagnostic accuracy and delivers improved saliency explanations.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Twin Network
Glaucoma Severity Assessment
Comparative System
🔎 Similar Papers
No similar papers found.
