Classification Tree-Based Machine Learning to Visualize and Validate a Decision Tool for Identifying Malnutrition in Cancer Patients.

JPEN. Journal of parenteral and enteral nutrition Pub Date : 2021-11-01 Epub Date: 2021-03-15 DOI:10.1002/jpen.2070
Liangyu Yin, Xin Lin, Jie Liu, Na Li, Xiumei He, Mengyuan Zhang, Jing Guo, Jian Yang, Li Deng, Yizhuo Wang, Tingting Liang, Chang Wang, Hua Jiang, Zhenming Fu, Suyi Li, Kunhua Wang, Zengqing Guo, Yi Ba, Wei Li, Chunhua Song, Jiuwei Cui, Hanping Shi, Hongxia Xu
{"title":"Classification Tree-Based Machine Learning to Visualize and Validate a Decision Tool for Identifying Malnutrition in Cancer Patients.","authors":"Liangyu Yin,&nbsp;Xin Lin,&nbsp;Jie Liu,&nbsp;Na Li,&nbsp;Xiumei He,&nbsp;Mengyuan Zhang,&nbsp;Jing Guo,&nbsp;Jian Yang,&nbsp;Li Deng,&nbsp;Yizhuo Wang,&nbsp;Tingting Liang,&nbsp;Chang Wang,&nbsp;Hua Jiang,&nbsp;Zhenming Fu,&nbsp;Suyi Li,&nbsp;Kunhua Wang,&nbsp;Zengqing Guo,&nbsp;Yi Ba,&nbsp;Wei Li,&nbsp;Chunhua Song,&nbsp;Jiuwei Cui,&nbsp;Hanping Shi,&nbsp;Hongxia Xu","doi":"10.1002/jpen.2070","DOIUrl":null,"url":null,"abstract":"<p><strong>Background: </strong>The newly proposed Global Leadership Initiative on Malnutrition (GLIM) framework is promising to gain global acceptance for diagnosing malnutrition. However, the role of machine learning in facilitating its application in clinical practice remains largely unknown.</p><p><strong>Methods: </strong>We performed a multicenter, observational cohort study including 3998 patients with cancer. Baseline malnutrition was defined using the GLIM criteria, and the study population was randomly divided into a derivation group (n = 2998) and a validation group (n = 1000). A classification and regression trees (CART) algorithm was used to develop a decision tree for classifying the severity of malnutrition in the derivation group. Model performance was evaluated in the validation group.</p><p><strong>Results: </strong>GLIM criteria diagnosed 588 patients (14.7%) with moderate malnutrition and 532 patients (13.3%) with severe malnutrition among the study population. The CART cross-validation identified 5 key predictors for the decision tree construction, including age, weight loss within 6 months, body mass index, calf circumference, and the Nutritional Risk Screening 2002 score. The decision tree showed high performance, with an area under the curve of 0.964 (κ = 0.898, P < .001, accuracy = 0.955) in the validation group. Subgroup analysis showed that the model had apparently good performance in different cancers. Among the 5 predictors constituting the tree, age contributed the least to the classification power.</p><p><strong>Conclusion: </strong>Using the machine learning, we visualized and validated a decision tool based on the GLIM criteria that can be conveniently used to accelerate the pretreatment identification of malnutrition in patients with cancer.</p>","PeriodicalId":520701,"journal":{"name":"JPEN. Journal of parenteral and enteral nutrition","volume":" ","pages":"1736-1748"},"PeriodicalIF":0.0000,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1002/jpen.2070","citationCount":"15","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"JPEN. Journal of parenteral and enteral nutrition","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1002/jpen.2070","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2021/3/15 0:00:00","PubModel":"Epub","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 15

Abstract

Background: The newly proposed Global Leadership Initiative on Malnutrition (GLIM) framework is promising to gain global acceptance for diagnosing malnutrition. However, the role of machine learning in facilitating its application in clinical practice remains largely unknown.

Methods: We performed a multicenter, observational cohort study including 3998 patients with cancer. Baseline malnutrition was defined using the GLIM criteria, and the study population was randomly divided into a derivation group (n = 2998) and a validation group (n = 1000). A classification and regression trees (CART) algorithm was used to develop a decision tree for classifying the severity of malnutrition in the derivation group. Model performance was evaluated in the validation group.

Results: GLIM criteria diagnosed 588 patients (14.7%) with moderate malnutrition and 532 patients (13.3%) with severe malnutrition among the study population. The CART cross-validation identified 5 key predictors for the decision tree construction, including age, weight loss within 6 months, body mass index, calf circumference, and the Nutritional Risk Screening 2002 score. The decision tree showed high performance, with an area under the curve of 0.964 (κ = 0.898, P < .001, accuracy = 0.955) in the validation group. Subgroup analysis showed that the model had apparently good performance in different cancers. Among the 5 predictors constituting the tree, age contributed the least to the classification power.

Conclusion: Using the machine learning, we visualized and validated a decision tool based on the GLIM criteria that can be conveniently used to accelerate the pretreatment identification of malnutrition in patients with cancer.

基于分类树的机器学习可视化和验证识别癌症患者营养不良的决策工具。
背景:新提出的营养不良全球领导倡议(GLIM)框架有望在诊断营养不良方面获得全球认可。然而,机器学习在促进其在临床实践中的应用中的作用在很大程度上仍然未知。方法:我们进行了一项包括3998例癌症患者的多中心、观察性队列研究。使用GLIM标准定义基线营养不良,并将研究人群随机分为衍生组(n = 2998)和验证组(n = 1000)。采用分类回归树(CART)算法建立决策树,对衍生组的营养不良严重程度进行分类。验证组对模型性能进行评价。结果:在研究人群中,GLIM标准诊断出588例(14.7%)中度营养不良和532例(13.3%)重度营养不良。CART交叉验证确定了决策树构建的5个关键预测因素,包括年龄、6个月内体重减轻、体重指数、小腿围和2002年营养风险筛查评分。在验证组中,决策树的曲线下面积为0.964 (κ = 0.898, P < 0.001,准确率= 0.955)。亚组分析表明,该模型在不同的肿瘤中均有明显的良好表现。在构成树的5个预测因子中,年龄对分类能力的贡献最小。结论:利用机器学习,我们可视化并验证了一个基于GLIM标准的决策工具,可以方便地用于加速癌症患者营养不良的预处理识别。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信