Related Experiment Video
Updated: Jun 29, 2026
08:10
An Ex Vivo Tissue Culture Model for Fibrovascular Complications in Proliferative Diabetic Retinopathy
Published on: January 25, 2019
7.9K
Risk factor analysis and clinical decision tree model construction for diabetic retinopathy in Western China
Yuan-Yuan Zhou1, Tai-Cheng Zhou2, Nan Chen3
1Department of Endocrinology and Metabolism, The Sixth Affiliated Hospital of Kunming Medical University, The People's Hospital of Yuxi City, Yuxi 653100, Yunnan Province, China.
World Journal of Diabetes
|November 28, 2022
Summary
Diabetic retinopathy (DR) is a leading cause of blindness in type 2 diabetes. A new decision tree model identifies key risk factors like diabetes duration and blood pressure for early DR prediction and prevention.
Area of Science:
- Ophthalmology
- Endocrinology
- Public Health
Background:
- Diabetic retinopathy (DR) is a major cause of blindness in type 2 diabetes mellitus (T2DM).
- Effective prevention and treatment strategies are lacking, highlighting the need for early detection.
- Yunnan province, China, faces a high prevalence of DR and economic challenges.
Purpose of the Study:
- To develop a clinical prediction model for early prevention and treatment of DR.
- To identify key risk factors for DR in the T2DM population.
Main Methods:
- A cross-sectional study of 1654 Han Chinese individuals with T2DM.
- Participants were categorized based on fundus photography: without DR (n=826) and with DR (n=828).
- Logistic regression and a clinical decision tree model were employed for analysis.
Main Results:
- Key risk factors identified include diabetes duration ≥ 10 years, female sex, elevated systolic blood pressure (SBP) (≥ 140 mmHg), and high cholesterol (≥ 6.22 mmol/L).
- Increased chronic kidney disease (CKD) severity and higher hemoglobin A1c levels also elevated DR risk.
- The decision tree model prioritized diabetes duration, CKD stage, and SBP (both supine and standing) as primary predictors.
Conclusions:
- A simple decision tree model effectively classifies DR risk.
- The model utilizes easily obtainable clinical data: diabetes duration, CKD stage, SBP, and BMI.
- This tool facilitates early DR risk assessment and management in T2DM patients.
