Research on a Diabetic Retinopathy Risk Prediction Model Integrating Health Profiles and Machine LearningSu Feiyu1, Feng Qishun2, Zeng Qingjia1, Gao Shuyao3, Xi Huinan1, Wu Wenxin1, Wang Yingshuai1, Hu Hongpu1

Corresponding author: HuHongpu, hu.hongpu@imicams.ac.cn

DOI: 10.12201/bmr.202601.00077

•  

  Abstract: Objective/Significance To construct a machine learning model integrated with health profiles for precise stratification and early identification of diabetic retinopathy risk, thereby providing personalized health management for patients. Methods/Process Based on a diabetic complication early warning dataset, the K-Means method was used to establish health profiles. A baseline linear discriminant analysis model was developed on the full sample. Seventeen machine learning algorithms, including logistic regression, linear discriminant analysis, K-nearest neighbors, and decision trees, were employed for modeling and compared with the baseline. The optimal prediction model for each profile was selected, and performance was evaluated using metrics such as accuracy, AUC, and F1 score. Model interpretability was analyzed using SHAP values.Results/Conclusion Five types of health profiles were successfully identified. The optimal models for most profiles outperformed the overall baseline in terms of AUC and accuracy, providing a feasible approach for improving the precision of chronic disease complication screening and personalized patient management.

  Key words: Diabetic retinopathy; Health profile; Machine learning; Risk prediction model

  Submit time: 27 January 2026

  Copyright: The copyright holder for this preprint is the author/funder, who has granted biomedRxiv a license to display the preprint in perpetuity.
• html

• 图表

• Cheng Luping, Wu Siyang, Lu Bo. Analysis of Risk Factors and Prediction of Type 2 Diabetes Mellitus Based on Machine Learning. 2025. doi: 10.12201/bmr.202509.00003

  LIU Yuehong, LU Xuan, DING Chaofei, JIN Fubi, CHEN Aiai, SUN Honger. Research progress on health education of patients with diabetic retinopathy. 2025. doi: 10.12201/bmr.202507.00049

  ruichen. Comparative and Explanatory Analysis of Influencing Factor Models for Type 2 Diabetes Mellitus Complicated with Hypertension Based on Machine Learning. 2025. doi: 10.12201/bmr.202511.00058

  guolianmei, sunxiaohong, zhangzhe, tianye. Construction of a Prediction Model for Lower Limb Deep Vein Thrombosis in Patients with Hemorrhagic Stroke Based on Machine Learning. 2025. doi: 10.12201/bmr.202510.00023

  wangyifan, shichaojun, maanning. Comparison of risk prediction models for atherosclerosis in type 2 diabetes mellitus. 2024. doi: 10.12201/bmr.202404.00007

  Mo Wei, Xiang Ya, Liao Qiujiao, He Liu, Ling Chaoling, Lu Qixiang, Liu Fangyin. Research progress on risk prediction model of postoperative delirium in elderly patients with hip fractureWEI Yunshi1? MO Wei1? XIANG YA1? LIAO Qiujiao1? HE Liu2? LING Chaoling2? LU Qixiang2? LIU Fangyin3▲. 2024. doi: 10.12201/bmr.202409.00029

  fan chun, ma jie, xu anqi, xu yihan, shi jinlong. Data Model Design and Intelligent Applications of Health Profiling. 2024. doi: 10.12201/bmr.202402.00011

  HONG Suru, CHEN Yushuang, WU Xiayang. Mortality risk assessment and interpretability analysis of preterm infants in the NICU using machine learning models. 2025. doi: 10.12201/bmr.202503.00066

  Wang Yan, ZHAO YONG. Mechanisms of Dietary Inflammation Modulation in Improving Diabetic Retinopathy. 2025. doi: 10.12201/bmr.202507.00069

  zhaochengyu. Research Progress on the Relationship between Hyperuricemia and Diabetic Peripheral Neuropathy. 2024. doi: 10.12201/bmr.202408.00019

• ID	Submit time	Number	Download
  1	2025-12-16	10.12201/bmr.202601.00077V1	Download

• Public  Anonymous  To author only Submit Comments