文献速递糖尿病

2型糖尿病患者慢性肾脏病多模态预测模型及其与血管并发症的关联:韩国与英国模型开发与验证研究

A Multimodal Predictive Model for Chronic Kidney Disease and Its Association With Vascular Complications in Patients With Type 2 Diabetes: Model Development and Validation Study in South Korea and the U.K.
PubMed 文献 · Diabetes Care · 2025-01-01

研究利用庆熙大学医疗中心7028例发现队列和英国生物银行1544例外部队列,开发并验证整合眼底图像与临床数据的多模态深度学习模型预测2型糖尿病患者5年内新发慢性肾脏病,模型在发现队列AUC为0.880(95%CI 0.806-0.954),外部队列为0.722,模型概率与血管并发症风险升高相关。

要点

  • 多模态模型整合视觉几何组16和深度神经网络,发现队列预测CKD的AUC为0.880(95%CI 0.806-0.954),外部队列为0.722。
  • SHAP和Grad-CAM分别突出eGFR和视盘为关键预测因子。
  • 模型概率与主要血管并发症风险升高相关,T1调整HR 1.42(95%CI 1.06-1.90),T2为1.59(1.17-2.16),T3为1.64(1.20-2.26);微血管并发症T3为1.30(1.02-1.67)。

核心结论

  • 多模态人工智能模型整合眼底图像和临床数据,可预测2型糖尿病患者5年内新发慢性肾脏病风险及相关的血管并发症。

研究设计与样本

  • 研究从庆熙大学医疗中心获取多模态数据集(n=7,028;发现队列)用于训练和内部验证,从英国生物银行获取数据集(n=1,544;验证队列)用于外部验证。
  • 慢性肾脏病基于ICD-9和ICD-10编码和/或估算肾小球滤过率(eGFR)≤60 mL/min/1.73 m2定义。
  • 研究集成多种深度学习模型,并使用可解释人工智能方法(包括Shapley加性解释值和梯度加权类激活映射)解释预测。

主要发现

  • 集成视觉几何组16和深度神经网络的多模态模型在发现队列预测CKD的受试者工作特征曲线下面积为0.880(95%CI 0.806-0.954),在验证队列为0.722。
  • SHAP和Grad-CAM分别突出eGFR和视盘为关键预测因子。
  • 模型概率与主要血管并发症风险升高相关:三分位1(T1)调整风险比1.42(95%CI 1.06-1.90);T2为1.59(1.17-2.16);T3为1.64(1.20-2.26)。
  • 模型概率与微血管并发症风险升高相关:T3调整风险比1.30(95%CI 1.02-1.67)。

局限性

  • 原文未报告局限性。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

原文

OBJECTIVE: To develop a multimodal model to predict chronic kidney disease (CKD) in patients with type 2 diabetes mellitus (T2DM), given the limited research on this integrative approach. RESEARCH DESIGN AND METHODS: We obtained multimodal data sets from Kyung Hee University Medical Center (n = 7,028; discovery cohort) for training and internal validation and UK Biobank (n = 1,544; validation cohort) for external validation. CKD was defined based on ICD-9 and ICD-10 codes and/or estimated glomerular filtration rate (eGFR) ≤60 mL/min/1.73 m2. We ensembled various deep learning models and interpreted their predictions using explainable artificial intelligence (AI) methods, including Shapley additive explanation values (SHAP) and gradient-weighted class activation mapping (Grad-CAM). Subsequently, we investigated the potential association between the model probability and vascular complications. RESULTS: The multimodal model, which ensembles visual geometry group 16 and deep neural network, presented high performance in predicting CKD, with area under the receiver operating characteristic curve of 0.880 (95% CI 0.806-0.954) in the discovery cohort and 0.722 in the validation cohort. SHAP and Grad-CAM highlighted key predictors, including eGFR and optic disc, respectively. The model probability was associated with an increased risk of macrovascular complications (tertile 1 [T1]: adjusted hazard ratio, 1.42 [95% CI 1.06-1.90]; T2: 1.59 [1.17-2.16]; T3: 1.64 [1.20-2.26]) and microvascular complications (T3: 1.30 [1.02-1.67]). CONCLUSIONS: Our multimodal AI model integrates fundus images and clinical data from binational cohorts to predict the risk of new-onset CKD within 5 years and associated vascular complications in patients with T2DM.

来源:Diabetes Care,查看原文。版权归原作者所有。

引用

  1. Cho J, Woo S, Hwang SH, et al. A Multimodal Predictive Model for Chronic Kidney Disease and Its Association With Vascular Complications in Patients With Type 2 Diabetes: Model Development and Validation Study in South Korea and the U.K. Diabetes Care. 2025. PMID: 40590639. doi:10.2337/dc25-0355

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。