PDF(Pubmed)
Abstract：
Older patients are at an increased risk of malnutrition due to many factors related to poor clinical outcomes.
This study aims to develop an assisted diagnosis model using machine learning (ML) for identifying older patients with malnutrition and providing the focus of individualized treatment.
We reanalyzed a multicenter, observational cohort study including 2660 older patients. Baseline malnutrition was defined using the global leadership initiative on malnutrition (GLIM) criteria, and the study population was randomly divided into a derivation group (2128/2660, 80%) and a validation group (532/2660, 20%). We applied 5 ML algorithms and further explored the relationship between features and the risk of malnutrition by using the Shapley additive explanations visualization method.
The proposed ML models were capable to identify older patients with malnutrition. In the external validation cohort, the top 3 models by the area under the receiver operating characteristic curve were light gradient boosting machine (92.1%), extreme gradient boosting (91.9%), and the random forest model (91.5%). Additionally, the analysis of the importance of features revealed that BMI, weight loss, and calf circumference were the strongest predictors to affect GLIM. A BMI of below 21 kg/m2 was associated with a higher risk of GLIM in older people.
We developed ML models for assisting diagnosis of malnutrition based on the GLIM criteria. The cutoff values of laboratory tests generated by Shapley additive explanations could provide references for the identification of malnutrition.
Chinese Clinical Trial Registry ChiCTR-EPC-14005253; https://www.chictr.org.cn/showproj.aspx?proj=9542.
This study aims to develop an assisted diagnosis model using machine learning (ML) for identifying older patients with malnutrition and providing the focus of individualized treatment.
We reanalyzed a multicenter, observational cohort study including 2660 older patients. Baseline malnutrition was defined using the global leadership initiative on malnutrition (GLIM) criteria, and the study population was randomly divided into a derivation group (2128/2660, 80%) and a validation group (532/2660, 20%). We applied 5 ML algorithms and further explored the relationship between features and the risk of malnutrition by using the Shapley additive explanations visualization method.
The proposed ML models were capable to identify older patients with malnutrition. In the external validation cohort, the top 3 models by the area under the receiver operating characteristic curve were light gradient boosting machine (92.1%), extreme gradient boosting (91.9%), and the random forest model (91.5%). Additionally, the analysis of the importance of features revealed that BMI, weight loss, and calf circumference were the strongest predictors to affect GLIM. A BMI of below 21 kg/m2 was associated with a higher risk of GLIM in older people.
We developed ML models for assisting diagnosis of malnutrition based on the GLIM criteria. The cutoff values of laboratory tests generated by Shapley additive explanations could provide references for the identification of malnutrition.
Chinese Clinical Trial Registry ChiCTR-EPC-14005253; https://www.chictr.org.cn/showproj.aspx?proj=9542.
摘要：
背景：由于许多与不良临床结局相关的因素，老年患者营养不良的风险增加。
目的：本研究旨在开发一种使用机器学习（ML）的辅助诊断模型，以识别营养不良的老年患者并提供个性化治疗的重点。
方法：我们重新分析了多中心，观察性队列研究，包括2660例老年患者。基线营养不良是使用全球营养不良领导倡议(GLIM)标准定义的，将研究人群随机分为推导组(2128/2660,80%)和验证组(532/2660,20%).我们应用了5种ML算法，并通过使用Shapley加性解释可视化方法进一步探索了特征与营养不良风险之间的关系。
结果：所提出的ML模型能够识别患有营养不良的老年患者。在外部验证队列中，按接收器工作特性曲线下面积计算，前3个型号为光梯度增强机(92.1%)，极端梯度提升（91.9%），和随机森林模型(91.5%)。此外,对特征重要性的分析表明，BMI，减肥,小腿围是影响GLIM的最强预测因子。BMI低于21kg/m2与老年人的GLIM风险较高相关。
结论：我们基于GLIM标准开发了辅助营养不良诊断的ML模型。通过Shapley添加剂解释产生的实验室测试的截止值可以为识别营养不良提供参考。
背景：中国临床试验注册中心ChiCTR-EPC-14005253;https://www.chictr.org.cn/showproj.aspx?proj=9542。
目的：本研究旨在开发一种使用机器学习（ML）的辅助诊断模型，以识别营养不良的老年患者并提供个性化治疗的重点。
方法：我们重新分析了多中心，观察性队列研究，包括2660例老年患者。基线营养不良是使用全球营养不良领导倡议(GLIM)标准定义的，将研究人群随机分为推导组(2128/2660,80%)和验证组(532/2660,20%).我们应用了5种ML算法，并通过使用Shapley加性解释可视化方法进一步探索了特征与营养不良风险之间的关系。
结果：所提出的ML模型能够识别患有营养不良的老年患者。在外部验证队列中，按接收器工作特性曲线下面积计算，前3个型号为光梯度增强机(92.1%)，极端梯度提升（91.9%），和随机森林模型(91.5%)。此外,对特征重要性的分析表明，BMI，减肥,小腿围是影响GLIM的最强预测因子。BMI低于21kg/m2与老年人的GLIM风险较高相关。
结论：我们基于GLIM标准开发了辅助营养不良诊断的ML模型。通过Shapley添加剂解释产生的实验室测试的截止值可以为识别营养不良提供参考。
背景：中国临床试验注册中心ChiCTR-EPC-14005253;https://www.chictr.org.cn/showproj.aspx?proj=9542。
