Department of Endocrinology, Dalian Municipal Central Hospital, Dalian, China.
Infervision Institute of Research, Beijing, China.
Front Endocrinol (Lausanne). 2022 May 17;13:876559. doi: 10.3389/fendo.2022.876559. eCollection 2022.
To construct and validate prediction models for the risk of diabetic retinopathy (DR) in patients with type 2 diabetes mellitus.
Patients with type 2 diabetes mellitus hospitalized over the period between January 2010 and September 2018 were retrospectively collected. Eighteen baseline demographic and clinical characteristics were used as predictors to train five machine-learning models. The model that showed favorable predictive efficacy was evaluated at annual follow-ups. Multi-point data of the patients in the test set were utilized to further evaluate the model's performance. We also assessed the relative prognostic importance of the selected risk factors for DR outcomes.
Of 7943 collected patients, 1692 (21.30%) developed DR during follow-up. Among the five models, the XGBoost model achieved the highest predictive performance with an AUC, accuracy, sensitivity, and specificity of 0.803, 88.9%, 74.0%, and 81.1%, respectively. The XGBoost model's AUCs in the different follow-up periods were 0.834 to 0.966. In addition to the classical risk factors of DR, serum uric acid (SUA), low-density lipoprotein cholesterol (LDL-C), total cholesterol (TC), estimated glomerular filtration rate (eGFR), and triglyceride (TG) were also identified to be important and strong predictors for the disease. Compared with the clinical diagnosis method of DR, the XGBoost model achieved an average of 2.895 years prior to the first diagnosis.
The proposed model achieved high performance in predicting the risk of DR among patients with type 2 diabetes mellitus at each time point. This study established the potential of the XGBoost model to facilitate clinicians in identifying high-risk patients and making type 2 diabetes management-related decisions.
构建和验证 2 型糖尿病患者糖尿病视网膜病变（DR）风险的预测模型。
回顾性收集 2010 年 1 月至 2018 年 9 月期间住院的 2 型糖尿病患者。将 18 项基线人口统计学和临床特征作为预测因子，用于训练 5 个机器学习模型。评估表现出良好预测效果的模型在年度随访中的表现。利用测试集中患者的多点数据进一步评估模型的性能。我们还评估了所选 DR 结局风险因素的相对预后重要性。
在收集的 7943 例患者中，1692 例（21.30%）在随访期间发生了 DR。在 5 个模型中，XGBoost 模型的预测性能最高，AUC、准确率、敏感度和特异度分别为 0.803、88.9%、74.0%和 81.1%。XGBoost 模型在不同随访期间的 AUC 为 0.834 至 0.966。除了 DR 的经典危险因素外，血清尿酸（SUA）、低密度脂蛋白胆固醇（LDL-C）、总胆固醇（TC）、估计肾小球滤过率（eGFR）和甘油三酯（TG）也被确定为疾病的重要和强预测因子。与 DR 的临床诊断方法相比，XGBoost 模型平均提前 2.895 年做出诊断。
所提出的模型在预测 2 型糖尿病患者 DR 风险方面在各个时间点均具有较高的性能。本研究确立了 XGBoost 模型在帮助临床医生识别高危患者和做出 2 型糖尿病管理相关决策方面的潜力。
