Diabetic Retinopathy Grading with CLIP-based Ranking-Aware Adaptation:A Comparative Study on Fundus Image
This work addresses the problem of large-scale screening for diabetic retinopathy, a leading cause of preventable blindness, by improving grading accuracy, though it is incremental as it builds on existing CLIP-based approaches.
The study tackled automated diabetic retinopathy severity grading by comparing three CLIP-based methods, finding that a ranking-aware model achieved the highest accuracy of 93.42% and AUROC of 0.9845, while a hybrid model excelled in detecting proliferative cases, both significantly outperforming a zero-shot baseline.
Diabetic retinopathy (DR) is a leading cause of preventable blindness, and automated fundus image grading can play an important role in large-scale screening. In this work, we investigate three CLIP-based approaches for five-class DR severity grading: (1) a zero-shot baseline using prompt engineering, (2) a hybrid FCN-CLIP model augmented with CBAM attention, and (3) a ranking-aware prompting model that encodes the ordinal structure of DR progression. We train and evaluate on a combined dataset of APTOS 2019 and Messidor-2 (n=5,406), addressing class imbalance through resampling and class-specific optimal thresholding. Our experiments show that the ranking-aware model achieves the highest overall accuracy (93.42%, AUROC 0.9845) and strong recall on clinically critical severe cases, while the hybrid FCN-CLIP model (92.49%, AUROC 0.99) excels at detecting proliferative DR. Both substantially outperform the zero-shot baseline (55.17%, AUROC 0.75). We analyze the complementary strengths of each approach and discuss their practical implications for screening contexts.