Benchmark Accuracy Does Not Transfer: Domain Adaptive Deep Learning for Diabetic Retinopathy Detection in a Bangladeshi Clinical Setting
Journal of Imaging Informatics in Medicine · 12.11.2026
DOI: 10.1007/s10278-026-02454-5 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Imaging Informatics in Medicine, online erschienen am 12.11.2026. Verfasst von Riad Hossain und 3 weiteren Autor:innen. Beteiligte Einrichtung: School of Science, Engineering and Technology, Chittagong, Bangladesh. riad.h@eastdelta.edu.bd. Die Arbeit nennt 44 Literaturangaben.
Abstract (Originalsprache)
Deep learning could make diabetic retinopathy (DR) screening more accessible where ophthalmologists are scarce, but high benchmark accuracy does not guarantee reliable clinical performance. We investigate this gap using 3662 images from APTOS 2019 for source training and a balanced evaluation corpus of 500 fundus images (250 DR, 250 No-DR) from a Bangladeshi clinic, graded by a single experienced ophthalmologist. Source-trained models achieved 94.26-96.80% accuracy under tenfold cross-validation on APTOS, yet achieved only 49.4-50.6% (mean 50.0%) when applied directly to the clinical images. The resulting 44.46-46.85 percentage-point losses across all six architectures provide consistent quantitative evidence of severe domain shift. A full-cohort feature comparison yielded a classical-feature Fréchet distance of 133.49 and an MMD 2 of 0.3967 (permutation p = 0.0005 ). To recover target-domain performance, we evaluate an adaptation pipeline that integrates retinal-specific preprocessing, progressive fine-tuning with discriminative learning rates, soft-voting inference, and test-time augmentation. Aggregated out-of-fold evaluation produced 81.0% accuracy (95% Wilson CI 77.3-84.2%), 70.0% sensitivity, 92.0% specificity, 89.7% precision, and a 78.7% F1 score. Fine-tuning increased mean individual-model accuracy by 27.1 percentage points, whereas preprocessing contributed a descriptive 2.0-point increase. These findings quantify a substantial benchmark-to-clinic gap and show that target-domain adaptation can improve performance, while the remaining false-negative rate and single-clinic design preclude claims of clinical readiness.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- School of Science, Engineering and Technology, Chittagong, Bangladesh. riad.h@eastdelta.edu.bd.
- School of Science, Engineering and Technology, Chittagong, Bangladesh.
- Department of Computer Science and Engineering, Chittagong University of Engineering and Technology, Chittagong, Bangladesh.
- East Delta University, Chittagong, Bangladesh.
Zitieren
APA
Hossain, R., Mahin Abir, I., Banu, A., & Nazim Uddin, M. (2026). Benchmark Accuracy Does Not Transfer: Domain Adaptive Deep Learning for Diabetic Retinopathy Detection in a Bangladeshi Clinical Setting. Journal of Imaging Informatics in Medicine. https://doi.org/10.1007/s10278-026-02454-5
Vancouver
Hossain R, Mahin Abir I, Banu A, Nazim Uddin M. Benchmark Accuracy Does Not Transfer: Domain Adaptive Deep Learning for Diabetic Retinopathy Detection in a Bangladeshi Clinical Setting. Journal of Imaging Informatics in Medicine. 2026. doi:10.1007/s10278-026-02454-5
BibTeX
@article{Hossain2026,
author = {Hossain, Riad and Mahin Abir, Ittekhar and Banu, Ayesha and Nazim Uddin, Mohammed},
title = {Benchmark Accuracy Does Not Transfer: Domain Adaptive Deep Learning for Diabetic Retinopathy Detection in a Bangladeshi Clinical Setting},
journal = {Journal of Imaging Informatics in Medicine},
year = {2026},
doi = {10.1007/s10278-026-02454-5}
}
EUVIMED