Review of medical decision support systems for the diagnosis of diabetic retinopathy

Diabetic retinopathy (DR) is one of the most common and severe complications of diabetes mellitus, often leading to blindness. Given the high prevalence of the disease and the limited capacity of ophthalmology services, the implementation of automated systems for early detection and monitoring of DR is becoming increasingly important. This review focuses on current advances in the application of computer vision and artificial intelligence (AI) techniques for DR diagnosis. A systematic search was conducted, followed by an analysis of 31 studies describing various approaches, including convolutional neural networks, pathology segmentation methods for fundus images, hybrid algorithms, and mobile applications for screening. Key features, model architectures, sensitivity and specificity metrics, as well as datasets used, are discussed. Special attention is given to AI systems already integrated into clinical practice, such as IDx-DR and EyeArt, which have received regulatory approval. The review highlights the importance of model interpretability, training data diversity, and image standardization as critical factors for improving the generalizability and trustworthiness of AI systems in ophthalmology. A distinctive aspect of this work is its comprehensive coverage of both international and Russian developments, including an assessment of their integration potential within the Russian healthcare system. The review concludes that several technologies have reached a level of maturity suitable for clinical use, while emphasizing the need for further research to enhance the accuracy, robustness, and transparency of diagnostic algorithms.

1. [Dedov II, Shestakova MV, Vikulova OK, et al. Diabetes mellitus in Russian Federation: dynamics of epidemiological indicators according to the Federal Diabetes Register for the period 2010-2022. Diabetes Mellitus. 2023;26(2):104–123. (In Russ.) doi: https://doi.org/10.14341/DM13035

2. Clinical recommendations «Diabetes mellitus type 1 in adults». Approved by the Ministry of Health of the Russian Federation; 2022. (In Russ.)

3. Clinical recommendations «Diabetes mellitus type 2 in adults». Approved by the Ministry of Health of the Russian Federation; 2022. (In Russ.)

4. Rebrova OY. Life cycle of decision support systems as medical technologies. Physician and information technologies. 2020;(1):27–37. (In Russ.).

5. Joshi GD, Sivaswamy J. DrishtiCare: A Telescreening Platform for Diabetic Retinopathy Powered with Fundus Image Analysis. Journal of Diabetes Science and Technology. 2011;5(1):23–31. doi: https://doi.org/10.1177/193229681100500104

6. Andreev DA, Kamynina NA, Antsiferov EI, et al. Telemedicine screening for diabetic retinopathy using digital technology: foreign experience. City Healthcare. 2024;5(1):103–111. (In Russ.) doi: https://doi.org/10.47619/2713-2617.zm.2024.5.1.103-111

7. Ribeiro ML, Nunes SG, Cunha-Vaz JG. Microaneurysm turnover at the macula predicts risk of development of clinically significant macular edema in persons with mild nonproliferative diabetic retinopathy. Diabetes Care. 2013;36(5):1254–1259. doi: https://doi.org/10.2337/dc12-1491

8. Grzybowski A, Brona P, Lim G, et al. Artificial intelligence for diabetic retinopathy screening: a review. Eye (Lond). 2020;34(3):451–460. doi: https://doi.org/10.1038/s41433-019-0566-0

9. Straтбk Z, Penибk M, Veith M. Artificial intelligence in diabetic retinopathy screening. A review. Cesk Slov Oftalmol. 2021;77(5):224–231. doi: https://doi.org/10.31348/2021/6

10. Tufail A, Rudisill C, Egan C, et al. Automated diabetic retinopathy image assessment software: diagnostic accuracy and cost-effectiveness compared with human graders. Ophthalmology. 2017;124(3):343–351. doi: https://doi.org/10.1016/j.ophtha.2016.11.014

11. Walton OB, Garoon RB, Weng CY, et al. Evaluation of Automated Teleretinal Screening Program for Diabetic Retinopathy. JAMA Ophthalmol. 2016;134(2):204–209. doi: https://doi.org/10.1001/jamaophthalmol.2015.5083

12. Pavlov VG, Sidamonidze AL, Petrachkov DV. Current trends in diabetic retinopathy screening. Vestnik Oftalmologii. 2020;136(4):300–307. (In Russ.) doi: https://doi.org/10.17116/oftalma2020136042300

13. Nakayama LF, Zago Ribeiro L, Novaes F, et al. Artificial intelligence for telemedicine diabetic retinopathy screening: a review. Ann Med. 2023;55(2):2258149. doi: https://doi.org/10.1080/07853890.2023.2258149

14. Gulshan V, Peng L, Coram M, et al. Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs. JAMA. 2016;316(22):2402–2410. doi: https://doi.org/10.1001/jama.2016.17216

15. Szegedy C, Vanhoucke V, Ioffe S, et al. Rethinking the Inception Architecture for Computer Vision. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR); Las Vegas, NV, USA;2016:2818–2826. doi: https://doi.org/10.1109/CVPR.2016.308

16. Dutt S, Sivaraman A, Savoy F, et al. Insights into the growing popularity of artificial intelligence in ophthalmology. Indian J Ophthalmol. 2020;68(7):1339–1346. doi: https://doi.org/10.4103/ijo.IJO_1754_19

17. Ji Y, Chen N, Liu S, et al. Research Progress of Artificial Intelligence Image Analysis in Systemic Disease-Related Ophthalmopathy. Dis Markers. 2022;2022:3406890. doi: https://doi.org/10.1155/2022/3406890

18. Takahashi H, Tampo H, Arai Y, et al. Applying artificial intelligence to disease staging: Deep learning for improved staging of diabetic retinopathy. PLoS ONE. 2017;12(6):e0179790. doi: https://doi.org/10.1371/journal.pone.0179790

19. Katalevskaya EA, Katalevsky DY, Tyurikov MI, et al. Prospects for the use of artificial intelligence in the diagnosis and treatment of retinal diseases. RMJ. Clinical Ophthalmology. 2022;22(1):36–43. (In Russ.) doi: https://doi.org/10.32364/2311-7729-2022-22-1-36-43

20. Ting DSW, Cheung CY, Lim G, et al. Development and Validation of a Deep Learning System for Diabetic Retinopathy and Related Eye Diseases Using Retinal Images From Multiethnic Populations With Diabetes. JAMA. 2017;318(22):2211–2223. doi: https://doi.org/10.1001/jama.2017.18152

21. Abramoff MD, Lavin PT, Birch M, et al. Pivotal trial of an autonomous AI-based diagnostic system for detection of diabetic retinopathy in primary care offices. NPJ Digit Med. 2018;1:39. doi: https://doi.org/10.1038/s41746-018-0040-6

22. Keskinbora K, Gьven F. Artificial Intelligence and Ophthalmology. Turk J Ophthalmol. 2020;50(1):37–43. doi: https://doi.org/10.4274/tjo.galenos.2020.78989

23. Padhy SK, Takkar B, Chawla R, et al. Artificial intelligence in diabetic retinopathy: A natural step to the future. Indian J Ophthalmol. 2019;67(7):1004–1009. doi: https://doi.org/10.4103/ijo.IJO_1989_18

24. Romanova DV, Gritsyuk EM, Stepanova EA, et al. On the subject of development of an expert system for diagnosing retinopathy of prematurity. Sistemnaya integratsiya v zdravookhranenii. 2020;(2):35–42. (In Russ.)

25. Martusevich YA, Kobyakova OS, Lyutsko VV. Early digital screening for diabetic retinopathy (Literature review). Sovremennye problemy zdravookhraneniya i meditsinskoi statistiki. 2023;(4):887–914. (In Russ.) doi: https://doi.org/10.24412/2312-2935-2023-4-887-914

26. Ryabikin DV, Mishchenko YV, Turchaninov GE. Training neural networks for the analysis and processing of medical data and disease diagnosis. Universitetskaya nauka. 2023;(2):178–183. (In Russ.)

27. Klimontov VV, Berikov VB, Saik OV. Artificial intelligence in diabetology. Diabetes Mellitus. 2021;24(2):156–166. (In Russ.) doi: https://doi.org/10.14341/DM12665

28. Grzybowski A, Brona P. Analysis and Comparison of Two Artificial Intelligence Diabetic Retinopathy Screening Algorithms in a Pilot Study: IDx-DR and Retinalyze. J Clin Med. 2021;10(11):2352. doi: https://doi.org/10.3390/jcm10112352

29. De Fauw J, Ledsam JR, Romera-Paredes B, et al. Clinically applicable deep learning for diagnosis and referral in retinal disease. Nat Med. 2018;24(9):1342–1350. doi: https://doi.org/10.1038/s41591-018-0107-6

30. Ryazanova SV, Komkov AA, Mazaev VP. Prospects for medical technologies of artificial intelligence. Nauchnoe obozrenie. Meditsinskie nauki. 2022;(4):90–94. (In Russ.) doi: https://doi.org/10.17513/srms.1279

31. Myasnyankina OP, Pron’kin NN. Achievements and prospects of artificial intelligence in medicine. International Journal of Professional Science. 2021;(4):27–32. (In Russ.) doi: https://www.elibrary.ru/item.asp?id=46179278

32. Saeed E, Szymkowski M, Saeed K, et al. An Approach to Automatic Hard Exudate Detection in Retina Color Images by a Telemedicine System Based on the d-Eye Sensor and Image Processing Algorithms. Sensors (Basel). 2019;19(3):695. doi: https://doi.org/10.3390/s19030695

33. Malerbi FK, Andrade RE, Morales PH, et al. Diabetic Retinopathy Screening Using Artificial Intelligence and Handheld SmartphoneBased Retinal Camera. Journal of Diabetes Science and Technology. 2022;16(3):716–723. doi: https://doi.org/10.1177/1932296820985567

34. Fedorovich AA, Gorshkov AY, Korolev AI, et al. Smartphone in medicine — from a handbook to a diagnostic system. Review of the current state of the issue. Kardiovaskulyarnaya terapiya i profilaktika. 2022;21(9):66–74. (In Russ.) doi: https://doi.org/10.15829/1728-8800-2022-3298

35. Bhaskaranand M, Ramachandra C, Bhat S, et al. 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 Technol Ther. 2019;21(11):635–643. doi: https://doi.org/10.1089/dia.2019.0164

36. Ferro Desideri L, Rutigliani C, Corazza P, et al. The upcoming role of Artificial Intelligence (AI) for retinal and glaucomatous diseases. J Optom. 2022;15(Suppl 1):S50–S57. doi: https://doi.org/10.1016/j.optom.2022.08.001

37. Wewetzer L, Held LA, Steinhдuser J. Diagnostic performance of deep-learning-based screening methods for diabetic retinopathy in primary care-A meta-analysis. PLoS ONE. 2021;16(8):e0255034. doi: https://doi.org/10.1371/journal.pone.0255034

38. Wang X, Ju L, Zhao X, et al. Retinal abnormalities recognition using regional multitask learning. Medical Image Computing and Computer Assisted Intervention – MICCAI 2019. 22nd International Conference; Shenzhen, China; 2019:30–38. doi: https://doi.org/10.1007/978-3-030-32245-8_4

39. Playout C, Duval R, Cheriet F. A Novel Weakly Supervised Multitask Architecture for Retinal Lesions Segmentation on Fundus Images. IEEE Transactions on Medical Imaging. 2019;38(10):2434–2444. doi: https://doi.org/10.1109/TMI.2019.2906319

40. Arenas-Cavalli JT, Abarca I, Rojas-Contreras M, et al. Clinical validation of an artificial intelligence-based diabetic retinopathy screening tool for a national health system. Eye (Lond). 2022;36(1):78–85. doi: https://doi.org/10.1038/s41433-020-01366-0

41. Kim TN, Myers F, Reber C, et al. A Smartphone-Based Tool for Rapid, Portable, and Automated Wide-Field Retinal Imaging. Transl Vis Sci Technol. 2018;7(5):21. doi: https://doi.org/10.1167/tvst.7.5.21

42. Mamedov TK, Dzyuba DV, Narkevich AN. Recognition of diabetic retinopathy on digital fundus images using a residual convolutional neural network. Vestnik novykh meditsinskikh tekhnologii. 2021;28(3):73–76. (In Russ.) doi: https://doi.org/10.24412/1609-2163-2021-3-73-76

43. Dukhova MA, Usachev VA. Design and development of a system for recognizing retinal defects based on a neural network. DSPA: Voprosy primeneniya tsifrovoy obrabotki signalov. 2021;11(1):39–46. (In Russ.)

44. Neroev VV, Bragin AA, Zaitseva OV. Development of a prototype service for diagnosing diabetic retinopathy from fundus images using artificial intelligence methods. Natsional’noe zdravookhranenie. 2021;2(2):64–72. (In Russ.) doi: https://doi.org/10.47093/2713-069X.2021.2.2.64-72

45. Katalevskaya EA, Katalevsky DY, Tyurikov MI, et al. Segmentation algorithm for visual signs of diabetic retinopathy (DR) and diabetic macular edema (DME) on digital fundus photographs. Rossiiskii zhurnal telemeditsiny i elektronnogo zdravookhraneniya. 2021;7(4):17–27. (In Russ.) doi: https://doi.org/10.29188/2712-9217-2021-7-4-17-26

46. Dai L, Wu L, Li H, et al. A deep learning system for detecting diabetic retinopathy across the disease spectrum. Nat Commun. 2021;12(1):3242. doi: https://doi.org/10.1038/s41467-021-23458-5

47. Zhelev Z, Peters J, Rogers M, et al. Test accuracy of artificial intelligence-based grading of fundus images in diabetic retinopathy screening: A systematic review. J Med Screen. 2023;30(3):97–112. doi: https://doi.org/10.1177/09691413221144382

48. Dong L, He W, Zhang R, et al. Artificial Intelligence for Screening of Multiple Retinal and Optic Nerve Diseases. JAMA Netw Open. 2022;5(5):e229960. doi: https://doi.org/10.1001/jamanetworkopen.2022.9960

49. Mahmood MAI, Aktar N, Kader MF. A hybrid approach for diagnosing diabetic retinopathy from fundus image exploiting deep features. Heliyon. 2023;9(9):e19625. doi: https://doi.org/10.1016/j.heliyon.2023.e19625

50. Lukashevich MM. Neural network classifier for determining diabetic retinopathy from retinal images. Sistemnyi analiz i prikladnaya informatika. 2023;(1):25–34. (In Russ.)doi: https://doi.org/10.21122/2309-4923-2023-1-25-34

51. Dzyuba DV, Narkevich AN, Kurbanismailov RB. Use of multidimensional models for diagnosing retinal pathologies. Vestnik Biomeditsina i Sotsiologiya. 2023;8(1):33–38. (In Russ.) doi: https://doi.org/10.26787/nydha-2618-8783-2023-8-1-33-38

52. Prathiba V, Rajalakshmi R, Arulmalar S, et al. Accuracy of the smartphone-based nonmydriatic retinal camera in the detection of sight-threatening diabetic retinopathy. Indian J Ophthalmol. 2020;68(Suppl 1):S42. doi: https://doi.org/10.4103/ijo.IJO_1937_19