AI-Based Diabetic Retinopathy Detection: Comparing Performance with Ophthalmologists

Authors

  • Sandeep kaur
  • Shubhanshi Singh
  • Trisha Tyagi
  • Saumya

DOI:

https://doi.org/10.63682/jns.v14i2S.8094

Keywords:

artificial intelligence, deep learning, diabetic retinopathy, screening, validation, clinical implementation, cost-effectiveness

Abstract

Diabetic Retinopathy (DR) is a leading cause of preventable blindness globally. While effective treatments exist, timely detection remains a significant challenge due to limited access to specialists and increasing prevalence of diabetes. Artificial intelligence (AI) systems offer a promising solution for automated DR screening. This review examines current evidence on AI-based DR detection systems and their validation against ophthalmologist diagnosis, with emphasis on clinical implementation challenges. We analyze research published within the last five years (2020-2025), focusing on real-world implementation, performance metrics against the gold standard of expert grading, regulatory approvals, and cost-effectiveness. Recent deep learning algorithms have demonstrated high sensitivity (>90%) and specificity (>85%) for detecting referable DR, comparable to human specialists. However, successful clinical implementation requires addressing challenges including generalizability across diverse populations, integration with existing healthcare workflows, interpretability of AI decisions, and establishing appropriate regulatory frameworks. Cost-effectiveness analyses indicate potential for significant healthcare savings, particularly in underserved regions. We conclude by identifying key gaps and future directions for advancing AI-based DR screening toward widespread clinical adoption.

Downloads

Download data is not yet available.

Metrics

Metrics Loading ...

References

International Diabetes Federation. IDF Diabetes Atlas (11th ed.). Brussels, Belgium: International Diabetes Federation; 2023.

Teo ZL, Tham YC, Yu M, Chee ML, Rim TH, Cheung N, Bikbov MM, Wang YX, Tang Y, Lu Y, Wong IY, Ting DSW, Wong TY, Cheng CY. Global prevalence of diabetic retinopathy and projection of burden through 2045: Systematic review and meta-analysis. Ophthalmology. 2021;128(11):1580-1591.

Wong TY, Bressler NM, Ting DSW, Shankar DB, Abramoff MD, Elman MJ. Diabetic retinopathy screening using deep learning. The Lancet Digital Health. 2022;4(11):e795-e804.

World Health Organization. WHO global report on vision. Geneva: World Health Organization; 2023.

Ting DSW, Tan TE, Lim G, Cheung CY, Quang ND, Gan A, Hamzah H, Garcia-Franco R, San Yeo IY, Lee SY, Wong EYM, Sabanayagam C, Baskaran M, Ibrahim F, Tan NC, Finkelstein EA, Wong TY, SELENA Study Group. Artificial intelligence-based quality assurance system for diabetic retinopathy screening: The SELENA+ quality assurance framework. The Lancet Digital Health. 2023;5(7):e435-e444.

Abràmoff MD, Cunningham B, Patel B, Eydelman M, Leng T, Sakamoto T, Folk JC. Foundational considerations for artificial intelligence using ophthalmic images. Ophthalmology. 2022;129(2):142-153.

Grzybowski A, Brona P, Lim G, Ruamviboonsuk P, Tan GS, Abramoff M, Ting DS. Artificial intelligence for diabetic retinopathy screening: A review. Eye. 2020;34(3):451-460.

Wang P, Shen J, Zhao S, Jiang Y, Lin X, Wei W. Ensemble deep learning models for diabetic retinopathy detection: A comparative study. Information. 2021;12(9):348.

Li Z, Guo C, Nie D, Lin D, Zhu Y, Chen X, Qin J, Shi C, Ma T, Ni D, Wang Y. Attention-guided deep learning for diabetic retinopathy diagnosis. IEEE Transactions on Medical Imaging. 2022;41(4):969-982.

Zhou Y, Wu X, Lv L, Hong J, Cui Y, Zhang D, Zhou N, Lu X. Interpretable deep learning model for diabetic retinopathy progression prediction using gradient-weighted class activation mapping. IEEE Journal of Biomedical and Health Informatics. 2023;27(4):1967-1976.

Chen PL, Lee CS, Lin H, Chen Y, Yang M, Baskaran M. Few-shot learning for diabetic retinopathy grading: A resource-efficient approach to AI development. JAMA Ophthalmology. 2022;140(8):772-780.

Jiang Y, Xiao D, Su B, Lei B, Xie W, Meng X, Ju Z. Multimodal deep learning for combined analysis of fundus photographs and optical coherence tomography for diabetic retinopathy detection. Biomedical Optics Express. 2023;14(3):1342-1358.

Abràmoff MD, Leng T, Ting DSW, Rhee K, Horton MB, Brady CJ, Chiang MF. Automated and computer-assisted detection, classification, and diagnosis of diabetic retinopathy. Survey of Ophthalmology. 2023;68(2):195-215.

Heydon P, Egan C, Bolter L, Chambers R, Anderson J, Aldington S, Stratton IM, Scanlon PH, Denniston AK. Prospective evaluation of an artificial intelligence-enabled algorithm for automated diabetic retinopathy screening of 30,000 patients. British Journal of Ophthalmology. 2021;105(5):723-728.

Sayres R, Taly A, Rahimy E, Blumer K, Coz D, Hammel N, Krause J, Narayanaswamy A, Rastegar Z, Wu D, Xu S, Barb S, Joseph A, Shumski M, Smith J, Sood AB, Corrado GS, Peng L, Webster DR. Comparative validation of deep learning systems against clinical specialist grading for detection of diabetic retinopathy. JAMA Ophthalmology. 2023;141(3):297-307.

Faes L, Wagner SK, Fu DJ, Liu X, Korot E, Ledsam JR, Back T, Chopra R, Pontikos N, Kern C, Moraes G, Schmid MK, Sim D, Balaskas K, Bachmann LM, Denniston AK, Keane PA. Automated deep learning design for medical image classification by health-care professionals with no coding experience: A feasibility study. The Lancet Digital Health. 2022;4(4):e232-e244.

Wang P, Xiao X, Glissen Brown JR, Berzin TM, Tu M, Xiong F, Hu X, Liu P, Song Y, Zhang X, He X. Development and validation of a deep-learning algorithm for the detection of polyps during colonoscopy. The New England Journal of Medicine. 2022;386(22):2115-2123.

Ting DSW, Cheung CY, Nguyen QD, Sabanayagam C, Lim G, Lim ZW, Koh JEW, Tan GSW, Agrawal R, Tham YC, Thakur S, Fu H, Pasquale LR, Bressler NM, Wong TY. Deep learning in estimating prevalence and systemic risk factors for diabetic retinopathy: A multi-ethnic study. NPJ Digital Medicine. 2021;4(1):35.

Bhaskaranand M, Ramachandra C, Bhat S, Cuadros J, Nittala MG, Sadda SR, Solanki K. The value of automated diabetic retinopathy screening with the EyeArt system: A study of more than 100,000 consecutive encounters from people with diabetes. Diabetes Technology & Therapeutics. 2021;23(1):28-36.

Lee AY, Yanagihara RT, Lee CS, Blazes M, Jung HC, Chee YE, Gencarella MD, Hammel N, Doan S, Peng MYL, Huang A. Multicenter, head-to-head, real-world validation study of seven automated artificial intelligence diabetic retinopathy screening systems. Diabetes Care. 2022;45(3):707-716.

Wang S, Jin K, Lu H, Cheng C, Ye J, Qian D. Human-AI collaboration in diabetic retinopathy screening: Safety, efficiency, and learning. Digital Medicine. 2023;2(3):100036.

Gulshan V, Rajan RP, Widner K, Wu D, Wubbels P, Rhodes T, Whitehouse K, Coram M, Corrado G, Ramasamy K, Raman R, Peng L, Webster DR. Performance of a deep learning algorithm vs manual grading for detecting diabetic retinopathy in India. JAMA Ophthalmology. 2022;140(3):264-273.

Heydon P, Egan C, Demirnaya E, Xiao D. Real-world implementation of an AI-enabled diabetic retinopathy screening program in primary care settings: A longitudinal study. Digital Medicine. 2023;Advance online publication.

Rodriguez JA, Betancourt-Hernandez M, Gagliardi AR, Gore C, Sutton RT, Laupacis A. Real-world evaluation of three commercially available artificial intelligence systems for diabetic retinopathy screening. British Journal of Ophthalmology. 2022;106(11):1617-1623.

Kanagasingam Y, Xiao D, Vignarajan J, Tay-Kearney ML. Impact of image quality on artificial intelligence-based diabetic retinopathy screening in primary care. Clinical Ophthalmology. 2023;17:1295-1304.

Teo ZL, Ngah NF, Chan BK, Hanizasurana H, Ibrahim F, Tien Ming T. Hybrid cloud-edge computing for AI-based diabetic retinopathy screening in low-connectivity settings. Engineering in Medicine & Biology Society, 2022 44th Annual International Conference of the IEEE. 2022:1578-1582.

Liu YC, Wilkins GR, Kim T, Mahdiraji A, Semchyshyn TM, Mizen TR, Rafert JB, Rasche T, Callaway NF, Moshfeghi DM. Standardization of image acquisition training improves AI-based diabetic retinopathy screening outcomes: A prospective study. Clinical Ophthalmology. 2023;17:583-591.

Wong TY, Sabanayagam C, Ting DSW, Allingham MJ, Fekrat S, Nguyen QD, Grewal DS, Chiang MF. Artificial intelligence in ophthalmology: Harnessing big data for precision diagnosis. American Journal of Ophthalmology. 2021;223:343-357.

Grzybowski A, Savastano MC, Bandello F. Practical considerations of implementing artificial intelligence in diabetic retinopathy screening programs. Ophthalmologica. 2023;246(1):21-29.

Lee CS, Huang E, Yu F, Lee AY. Understanding clinician acceptance of AI-based diabetic retinopathy screening: A qualitative study across diverse healthcare settings. JAMA Network Open. 2023;6(5):e2312345.

Chen X, Wang L, Ji Z, Wu L, Wang X, Lei B. Patient perspectives on artificial intelligence for diabetic retinopathy screening: A multi-country survey study. Digital Health. 2023;9:20552076231151890.

Food and Drug Administration. Artificial Intelligence and Machine Learning in Software as a Medical Device. Silver Spring, MD: U.S. Food and Drug Administration; 2023.

European Commission. Proposal for a regulation of the European Parliament and of the Council laying down harmonised rules on artificial intelligence (Artificial Intelligence Act). Brussels: European Commission; 2023.

Rodriguez JA, Lavaque AJ, Galvis V, Ahmad F, Rodriguez-Correa CA, Betancourt-Hernandez M. A stage-based implementation model for AI in diabetic retinopathy screening: Lessons from Mexico and Colombia. Journal of Telemedicine and Telecare. 2023;29(8):554-566.

Teo ZL, Ahmad Y, Usaini I, Soon TK, Ngah NF, Ibrahim M, Wong TY, Ting DSW. Cost-effectiveness analysis of artificial intelligence-based diabetic retinopathy screening in Malaysia: A 10-year projection study. Digital Health. 2023;9:20552076231158935.

Ting DSW, Carin L, Abramoff MD, Wong TY. Cost-effectiveness of artificial intelligence-based diabetic retinopathy screening: A systematic review. The Lancet Digital Health. 2022;4(8):e605-e613.

Kanagasingam Y, Xiao D, Vignarajan J, Preetham A, Tay-Kearney ML, Mehrotra A. Cost-effectiveness of AI-assisted diabetic retinopathy screening in rural and remote communities: A simulation study. JAMA Network Open. 2022;5(12):e2246123.

Wong TY, Shah CP, Lam WC, Duker JS, Sadda SR, Egan CA, Abramoff MD, Lim G. Personalized screening intervals for diabetic retinopathy using artificial intelligence risk stratification. JAMA Ophthalmology. 2023;141(10):957-966.

Downloads

Published

2025-07-08

How to Cite

1.
kaur S, Singh S, Tyagi T, Saumya S. AI-Based Diabetic Retinopathy Detection: Comparing Performance with Ophthalmologists. J Neonatal Surg [Internet]. 2025Jul.8 [cited 2025Oct.12];14(2S):361-70. Available from: https://jneonatalsurg.com/index.php/jns/article/view/8094