Development and Validation of a Machine Learning Algorithm for Predicting Diabetes Retinopathy in Patients With Type 2 Diabetes: Algorithm Development Study

Overview

Journal JMIR Med Inform

Publisher JMIR Publications

Date 2025 Feb 9

PMID 39924304

Authors

Sunyoung Kim

Jaeyu Park

Yejun Son

Hojae Lee

Selin Woo

Myeongcheol Lee

Hayeon Lee

Hyunji Sang

Dong Keon Yon

Sang Youl Rhee

Affiliations

Soon will be listed here.

Abstract

Background: Diabetic retinopathy (DR) is the leading cause of preventable blindness worldwide. Machine learning (ML) systems can enhance DR in community-based screening. However, predictive power models for usability and performance are still being determined.

Objective: This study used data from 3 university hospitals in South Korea to conduct a simple and accurate assessment of ML-based risk prediction for the development of DR that can be universally applied to adults with type 2 diabetes mellitus (T2DM).

Methods: DR was predicted using data from 2 independent electronic medical records: a discovery cohort (one hospital, n=14,694) and a validation cohort (2 hospitals, n=1856). The primary outcome was the presence of DR at 3 years. Different ML-based models were selected through hyperparameter tuning in the discovery cohort, and the area under the receiver operating characteristic (ROC) curve was analyzed in both cohorts.

Results: Among 14,694 patients screened for inclusion, 348 (2.37%) were diagnosed with DR. For DR, the extreme gradient boosting (XGBoost) system had an accuracy of 75.13% (95% CI 74.10-76.17), a sensitivity of 71.00% (95% CI 66.83-75.17), and a specificity of 75.23% (95% CI 74.16-76.31) in the original dataset. Among the validation datasets, XGBoost had an accuracy of 65.14%, a sensitivity of 64.96%, and a specificity of 65.15%. The most common feature in the XGBoost model is dyslipidemia, followed by cancer, hypertension, chronic kidney disease, neuropathy, and cardiovascular disease.

Conclusions: This approach shows the potential to enhance patient outcomes by enabling timely interventions in patients with T2DM, improving our understanding of contributing factors, and reducing DR-related complications. The proposed prediction model is expected to be both competitive and cost-effective, particularly for primary care settings in South Korea.

Citing Articles

Circulating CCN6/WISP3 in type 2 diabetes mellitus patients and its correlation with insulin resistance and inflammation: statistical and machine learning analyses.

Afrisham R, Jadidi Y, Moradi N, Ayyoubzadeh S, Fadaei R, Kiani Ghalesardi O BMC Med Inform Decis Mak. 2025; 25(1):114.

PMID: 40050813 PMC: 11887242. DOI: 10.1186/s12911-025-02957-1.

References

Ting D, Yim-Lui Cheung C, Lim G, Wei Tan G, Quang N, Gan A . 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. PMC: 5820739. DOI: 10.1001/jama.2017.18152. View

Saeedi P, Petersohn I, Salpea P, Malanda B, Karuranga S, Unwin N . Global and regional diabetes prevalence estimates for 2019 and projections for 2030 and 2045: Results from the International Diabetes Federation Diabetes Atlas, 9 edition. Diabetes Res Clin Pract. 2019; 157:107843. DOI: 10.1016/j.diabres.2019.107843. View

Hong J, Surapaneni A, Daya N, Selvin E, Coresh J, Grams M . Retinopathy and Risk of Kidney Disease in Persons With Diabetes. Kidney Med. 2021; 3(5):808-815.e1. PMC: 8515075. DOI: 10.1016/j.xkme.2021.04.018. View

Zhou Y, Wang C, Shi K, Yin X . Relationship between dyslipidemia and diabetic retinopathy: A systematic review and meta-analysis. Medicine (Baltimore). 2018; 97(36):e12283. PMC: 6133445. DOI: 10.1097/MD.0000000000012283. View

Lima V, Cavalieri G, Lima M, Nazario N, Lima G . Risk factors for diabetic retinopathy: a case-control study. Int J Retina Vitreous. 2016; 2:21. PMC: 5088444. DOI: 10.1186/s40942-016-0047-6. View

Dai L, Wu L, Li H, Cai C, Wu Q, Kong H . A deep learning system for detecting diabetic retinopathy across the disease spectrum. Nat Commun. 2021; 12(1):3242. PMC: 8163820. DOI: 10.1038/s41467-021-23458-5. View

Jeng C, Hsieh Y, Yang C, Yang C, Lin C, Wang I . Diabetic Retinopathy in Patients with Dyslipidemia: Development and Progression. Ophthalmol Retina. 2019; 2(1):38-45. DOI: 10.1016/j.oret.2017.05.010. View

Rasheed R, Pillai G, Kumar H, Shajan A, Radhakrishnan N, Ravindran G . Relationship between diabetic retinopathy and diabetic peripheral neuropathy - Neurodegenerative and microvascular changes. Indian J Ophthalmol. 2021; 69(11):3370-3375. PMC: 8725091. DOI: 10.4103/ijo.IJO_1279_21. View

Gayathri S, Gopi V, Palanisamy P . Diabetic retinopathy classification based on multipath CNN and machine learning classifiers. Phys Eng Sci Med. 2021; 44(3):639-653. DOI: 10.1007/s13246-021-01012-3. View

10.

Faria J, Silva K, Lopes de Faria J . The contribution of hypertension to diabetic nephropathy and retinopathy: the role of inflammation and oxidative stress. Hypertens Res. 2011; 34(4):413-22. DOI: 10.1038/hr.2010.263. View

11.

Ruamviboonsuk P, Krause J, Chotcomwongse P, Sayres R, Raman R, Widner K . Erratum: Author Correction: Deep learning versus human graders for classifying diabetic retinopathy severity in a nationwide screening program. NPJ Digit Med. 2019; 2:68. PMC: 6650411. DOI: 10.1038/s41746-019-0146-5. View

12.

Liu H, Li R, Zhang Y, Zhang K, Yusufu M, Liu Y . Economic evaluation of combined population-based screening for multiple blindness-causing eye diseases in China: a cost-effectiveness analysis. Lancet Glob Health. 2023; 11(3):e456-e465. DOI: 10.1016/S2214-109X(22)00554-X. View

13.

Wacker C, Prkno A, Brunkhorst F, Schlattmann P . Procalcitonin as a diagnostic marker for sepsis: a systematic review and meta-analysis. Lancet Infect Dis. 2013; 13(5):426-35. DOI: 10.1016/S1473-3099(12)70323-7. View

14.

Hsu C, Lee C, Chou K, Lee C, Chen H, Chiou J . The Association of Diabetic Retinopathy and Cardiovascular Disease: A 13-Year Nationwide Population-Based Cohort Study. Int J Environ Res Public Health. 2021; 18(15). PMC: 8345672. DOI: 10.3390/ijerph18158106. View

15.

Wong T, Sabanayagam C . The War on Diabetic Retinopathy: Where Are We Now?. Asia Pac J Ophthalmol (Phila). 2019; 8(6):448-456. PMC: 6903323. DOI: 10.1097/APO.0000000000000267. View

16.

Daskivich L, Vasquez C, Martinez Jr C, Tseng C, Mangione C . Implementation and Evaluation of a Large-Scale Teleretinal Diabetic Retinopathy Screening Program in the Los Angeles County Department of Health Services. JAMA Intern Med. 2017; 177(5):642-649. PMC: 5818774. DOI: 10.1001/jamainternmed.2017.0204. View

17.

Alali N, Albazei A, Alotaibi H, Almohammadi A, Alsirhani E, Alanazi T . Diabetic Retinopathy and Eye Screening: Diabetic Patients Standpoint, Their Practice, and Barriers; A Cross-Sectional Study. J Clin Med. 2022; 11(21). PMC: 9654427. DOI: 10.3390/jcm11216351. View

18.

Eum S, Son J, Min C, Cho W, Kim S, Woo H . Ethnic and sex differences in the distributions of body mass index and waist circumference among adults: a binationally representative study in South Korea and the United States. Eur Rev Med Pharmacol Sci. 2023; 27(5):1889-1903. DOI: 10.26355/eurrev_202303_31555. View

19.

Stratton I, Kohner E, Aldington S, Turner R, Holman R, Manley S . UKPDS 50: risk factors for incidence and progression of retinopathy in Type II diabetes over 6 years from diagnosis. Diabetologia. 2001; 44(2):156-63. DOI: 10.1007/s001250051594. View

20.

Li Z, Keel S, Liu C, He Y, Meng W, Scheetz J . An Automated Grading System for Detection of Vision-Threatening Referable Diabetic Retinopathy on the Basis of Color Fundus Photographs. Diabetes Care. 2018; 41(12):2509-2516. DOI: 10.2337/dc18-0147. View