Hou Rui, Dou Jingtao, Wu Lijuan, Zhang Xiaoyu, Li Changwei, Wang Weiqing, Gao Zhengnan, Tang Xulei, Yan Li, Wan Qin, Luo Zuojie, Qin Guijun, Chen Lulu, Ji Jianguang, He Yan, Wang Wei, Mu Yiming, Zheng Deqiang
Department of Epidemiology and Health Statistics, School of Public Health, Capital Medical University, Beijing, China.
Beijing Center for Disease Prevention and Control, Beijing, China.
Diabetes Metab Res Rev. 2024 Jul;40(5):e3832. doi: 10.1002/dmrr.3832.
Due to the high cost and complexity, the oral glucose tolerance test is not adopted as the screening method for identifying diabetes patients, which leads to the misdiagnosis of patients with isolated post-challenge hyperglycemia (IPH), that is., patients with normal fasting plasma glucose (<7.0 mmoL/L) and abnormal 2-h postprandial blood glucose (≥11.1 mmoL/L). We aimed to develop a model to differentiate individuals with IPH from the normal population.
Data from 54301 eligible participants were obtained from the Risk Evaluation of Cancers in Chinese Diabetic Individuals: a longitudinal (REACTION) study in China. Data from 37740 participants were used to develop the diagnostic system. External validation was performed among 16561 participants. Three machine learning algorithms were used to create the predictive models, which were further evaluated by various classification algorithms to establish the best predictive model.
Ten features were selected to develop an IPH diagnosis system (IPHDS) based on an artificial neural network. In external validation, the AUC of the IPHDS was 0.823 (95% CI 0.811-0.836), which was significantly higher than the AUC of the Taiwan model [0.799 (0.786-0.813)] and that of the Chinese Diabetes Risk Score model [0.648 (0.635-0.662)]. The IPHDS model had a sensitivity of 75.6% and a specificity of 74.6%. This model outperformed the Taiwan and CDRS models in subgroup analyses. An online site with instant predictions was deployed at https://app-iphds-e1fc405c8a69.herokuapp.com/.
The proposed IPHDS could be a convenient and user-friendly screening tool for diabetes during health examinations in a large general population.
由于成本高且操作复杂，口服葡萄糖耐量试验未被用作糖尿病患者的筛查方法，这导致了单纯餐后高血糖（IPH）患者的误诊，即空腹血糖正常（<7.0mmol/L）而餐后2小时血糖异常（≥11.1mmol/L）的患者。我们旨在开发一种模型，以区分IPH个体与正常人群。
从中国糖尿病个体癌症风险评估：一项纵向（REACTION）研究中获取了54301名符合条件参与者的数据。37740名参与者的数据用于开发诊断系统。在16561名参与者中进行了外部验证。使用三种机器学习算法创建预测模型，并通过各种分类算法进一步评估，以建立最佳预测模型。
选择了10个特征来开发基于人工神经网络的IPH诊断系统（IPHDS）。在外部验证中，IPHDS的AUC为0.823（95%CI 0.811 - 0.836），显著高于台湾模型的AUC [0.799（0.786 - 0.813）]和中国糖尿病风险评分模型的AUC [0.648（0.635 - 0.662）]。IPHDS模型的敏感性为75.6%，特异性为74.6%。在亚组分析中，该模型优于台湾模型和CDRS模型。一个具有即时预测功能的在线网站已部署在https://app-iphds-e1fc405c8a69.herokuapp.com/ 。
所提出的IPHDS可以成为在广大普通人群健康检查期间用于糖尿病筛查的便捷且用户友好的工具。
