Development and validation of a machine learning-based model for post-sepsis frailty.

IF 4.3 3区 医学 Q1 RESPIRATORY SYSTEM
ERJ Open Research Pub Date : 2024-10-07 eCollection Date: 2024-09-01 DOI:10.1183/23120541.00166-2024
Hye Ju Yeo, Dasom Noh, Tae Hwa Kim, Jin Ho Jang, Young Seok Lee, Sunghoon Park, Jae Young Moon, Kyeongman Jeon, Dong Kyu Oh, Su Yeon Lee, Mi Hyeon Park, Chae-Man Lim, Woo Hyun Cho, Sunyoung Kwon
{"title":"Development and validation of a machine learning-based model for post-sepsis frailty.","authors":"Hye Ju Yeo, Dasom Noh, Tae Hwa Kim, Jin Ho Jang, Young Seok Lee, Sunghoon Park, Jae Young Moon, Kyeongman Jeon, Dong Kyu Oh, Su Yeon Lee, Mi Hyeon Park, Chae-Man Lim, Woo Hyun Cho, Sunyoung Kwon","doi":"10.1183/23120541.00166-2024","DOIUrl":null,"url":null,"abstract":"<p><strong>Background: </strong>The development of post-sepsis frailty is a common and significant problem, but it is a challenge to predict.</p><p><strong>Methods: </strong>Data for deep learning were extracted from a national multicentre prospective observational cohort of patients with sepsis in Korea between September 2019 and December 2021. The primary outcome was frailty at survival discharge, defined as a clinical frailty score on the Clinical Frailty Scale ≥5. We developed a deep learning model for predicting frailty after sepsis by 10 variables routinely collected at the recognition of sepsis. With cross-validation, we trained and tuned six machine learning models, including four conventional and two neural network models. Moreover, we computed the importance of each predictor variable in the model. We measured the performance of these models using a temporal validation data set.</p><p><strong>Results: </strong>A total of 8518 patients were included in the analysis; 5463 (64.1%) were frail, and 3055 (35.9%) were non-frail at discharge. The Extreme Gradient Boosting (XGB) achieved the highest area under the receiver operating characteristic curve (AUC) (0.8175) and accuracy (0.7414). To confirm the generalisation performance of artificial intelligence in predicting frailty at discharge, we conducted external validation with the COVID-19 data set. The XGB still showed a good performance with an AUC of 0.7668. The machine learning model could predict frailty despite the disparity in data distribution.</p><p><strong>Conclusion: </strong>The machine learning-based model developed for predicting frailty after sepsis achieved high performance with limited baseline clinical parameters.</p>","PeriodicalId":11739,"journal":{"name":"ERJ Open Research","volume":"10 5","pages":""},"PeriodicalIF":4.3000,"publicationDate":"2024-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11456972/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"ERJ Open Research","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1183/23120541.00166-2024","RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2024/9/1 0:00:00","PubModel":"eCollection","JCR":"Q1","JCRName":"RESPIRATORY SYSTEM","Score":null,"Total":0}
引用次数: 0

Abstract

Background: The development of post-sepsis frailty is a common and significant problem, but it is a challenge to predict.

Methods: Data for deep learning were extracted from a national multicentre prospective observational cohort of patients with sepsis in Korea between September 2019 and December 2021. The primary outcome was frailty at survival discharge, defined as a clinical frailty score on the Clinical Frailty Scale ≥5. We developed a deep learning model for predicting frailty after sepsis by 10 variables routinely collected at the recognition of sepsis. With cross-validation, we trained and tuned six machine learning models, including four conventional and two neural network models. Moreover, we computed the importance of each predictor variable in the model. We measured the performance of these models using a temporal validation data set.

Results: A total of 8518 patients were included in the analysis; 5463 (64.1%) were frail, and 3055 (35.9%) were non-frail at discharge. The Extreme Gradient Boosting (XGB) achieved the highest area under the receiver operating characteristic curve (AUC) (0.8175) and accuracy (0.7414). To confirm the generalisation performance of artificial intelligence in predicting frailty at discharge, we conducted external validation with the COVID-19 data set. The XGB still showed a good performance with an AUC of 0.7668. The machine learning model could predict frailty despite the disparity in data distribution.

Conclusion: The machine learning-based model developed for predicting frailty after sepsis achieved high performance with limited baseline clinical parameters.

开发并验证基于机器学习的败血症后虚弱模型。
背景:败血症后虚弱的发展是一个常见的重大问题,但要预测它却是一个挑战:从2019年9月至2021年12月期间韩国脓毒症患者的全国多中心前瞻性观察队列中提取数据进行深度学习。主要结果是生存出院时的虚弱程度,定义为临床虚弱量表的临床虚弱评分≥5。我们开发了一个深度学习模型,通过脓毒症识别时常规收集的 10 个变量来预测脓毒症后的虚弱程度。通过交叉验证,我们训练并调整了六个机器学习模型,包括四个传统模型和两个神经网络模型。此外,我们还计算了模型中每个预测变量的重要性。我们使用时间验证数据集测量了这些模型的性能:共有 8518 名患者被纳入分析,其中 5463 人(64.1%)出院时体弱,3055 人(35.9%)出院时非体弱。极端梯度提升法(XGB)的接收者工作特征曲线下面积(AUC)(0.8175)和准确率(0.7414)最高。为了证实人工智能在预测出院时虚弱程度方面的普适性能,我们使用 COVID-19 数据集进行了外部验证。XGB 的 AUC 为 0.7668,仍然表现出色。尽管数据分布存在差异,但机器学习模型仍能预测虚弱程度:结论:为预测脓毒症后的虚弱程度而开发的基于机器学习的模型在基线临床参数有限的情况下取得了很高的性能。
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
ERJ Open Research
ERJ Open Research Medicine-Pulmonary and Respiratory Medicine
CiteScore
6.20
自引率
4.30%
发文量
273
审稿时长
8 weeks
期刊介绍: ERJ Open Research is a fully open access original research journal, published online by the European Respiratory Society. The journal aims to publish high-quality work in all fields of respiratory science and medicine, covering basic science, clinical translational science and clinical medicine. The journal was created to help fulfil the ERS objective to disseminate scientific and educational material to its members and to the medical community, but also to provide researchers with an affordable open access specialty journal in which to publish their work.
×
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:481959085
Book学术官方微信