A Machine Learning–Based Prediction Model for Acute Kidney Injury in Patients With Community-Acquired Pneumonia: Multicenter Validation Study

BackgroundAcute kidney injury (AKI) is common in patients with community-acquired pneumonia (CAP) and is associated with increased morbidity and mortality. ObjectiveThis study aimed to establish and validate predictive models for AKI in hospitalized patients with...

Full description

Saved in:
Bibliographic Details
Main Authors: Mengqing Ma, Caimei Chen, Dawei Chen, Hao Zhang, Xia Du, Qing Sun, Li Fan, Huiping Kong, Xueting Chen, Changchun Cao, Xin Wan
Format: Article
Language:English
Published: JMIR Publications 2024-12-01
Series:Journal of Medical Internet Research
Online Access:https://www.jmir.org/2024/1/e51255
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850249164570492928
author Mengqing Ma
Caimei Chen
Dawei Chen
Hao Zhang
Xia Du
Qing Sun
Li Fan
Huiping Kong
Xueting Chen
Changchun Cao
Xin Wan
author_facet Mengqing Ma
Caimei Chen
Dawei Chen
Hao Zhang
Xia Du
Qing Sun
Li Fan
Huiping Kong
Xueting Chen
Changchun Cao
Xin Wan
author_sort Mengqing Ma
collection DOAJ
description BackgroundAcute kidney injury (AKI) is common in patients with community-acquired pneumonia (CAP) and is associated with increased morbidity and mortality. ObjectiveThis study aimed to establish and validate predictive models for AKI in hospitalized patients with CAP based on machine learning algorithms. MethodsWe trained and externally validated 5 machine learning algorithms, including logistic regression, support vector machine, random forest, extreme gradient boosting, and deep forest (DF). Feature selection was conducted using the sliding window forward feature selection technique. Shapley additive explanations and local interpretable model-agnostic explanation techniques were applied to the optimal model for visual interpretation. ResultsA total of 6371 patients with CAP met the inclusion criteria. The development of CAP-associated AKI (CAP-AKI) was recognized in 1006 (15.8%) patients. The 11 selected indicators were sex, temperature, breathing rate, diastolic blood pressure, C-reactive protein, albumin, white blood cell, hemoglobin, platelet, blood urea nitrogen, and neutrophil count. The DF model achieved the best area under the receiver operating characteristic curve (AUC) and accuracy in the internal (AUC=0.89, accuracy=0.90) and external validation sets (AUC=0.87, accuracy=0.83). Furthermore, the DF model had the best calibration among all models. In addition, a web-based prediction platform was developed to predict CAP-AKI. ConclusionsThe model described in this study is the first multicenter-validated AKI prediction model that accurately predicts CAP-AKI during hospitalization. The web-based prediction platform embedded with the DF model serves as a user-friendly tool for early identification of high-risk patients.
format Article
id doaj-art-1a92cd63f8244b0b92c1b1e6a751ae09
institution OA Journals
issn 1438-8871
language English
publishDate 2024-12-01
publisher JMIR Publications
record_format Article
series Journal of Medical Internet Research
spelling doaj-art-1a92cd63f8244b0b92c1b1e6a751ae092025-08-20T01:58:33ZengJMIR PublicationsJournal of Medical Internet Research1438-88712024-12-0126e5125510.2196/51255A Machine Learning–Based Prediction Model for Acute Kidney Injury in Patients With Community-Acquired Pneumonia: Multicenter Validation StudyMengqing Mahttps://orcid.org/0000-0003-4402-7988Caimei Chenhttps://orcid.org/0000-0003-4954-954XDawei Chenhttps://orcid.org/0000-0001-5088-6565Hao Zhanghttps://orcid.org/0000-0001-8055-7789Xia Duhttps://orcid.org/0009-0005-9090-6264Qing Sunhttps://orcid.org/0009-0007-2878-1194Li Fanhttps://orcid.org/0000-0002-7853-3195Huiping Konghttps://orcid.org/0000-0002-4384-1251Xueting Chenhttps://orcid.org/0009-0007-5593-9110Changchun Caohttps://orcid.org/0000-0003-1992-4257Xin Wanhttps://orcid.org/0000-0001-5226-4444 BackgroundAcute kidney injury (AKI) is common in patients with community-acquired pneumonia (CAP) and is associated with increased morbidity and mortality. ObjectiveThis study aimed to establish and validate predictive models for AKI in hospitalized patients with CAP based on machine learning algorithms. MethodsWe trained and externally validated 5 machine learning algorithms, including logistic regression, support vector machine, random forest, extreme gradient boosting, and deep forest (DF). Feature selection was conducted using the sliding window forward feature selection technique. Shapley additive explanations and local interpretable model-agnostic explanation techniques were applied to the optimal model for visual interpretation. ResultsA total of 6371 patients with CAP met the inclusion criteria. The development of CAP-associated AKI (CAP-AKI) was recognized in 1006 (15.8%) patients. The 11 selected indicators were sex, temperature, breathing rate, diastolic blood pressure, C-reactive protein, albumin, white blood cell, hemoglobin, platelet, blood urea nitrogen, and neutrophil count. The DF model achieved the best area under the receiver operating characteristic curve (AUC) and accuracy in the internal (AUC=0.89, accuracy=0.90) and external validation sets (AUC=0.87, accuracy=0.83). Furthermore, the DF model had the best calibration among all models. In addition, a web-based prediction platform was developed to predict CAP-AKI. ConclusionsThe model described in this study is the first multicenter-validated AKI prediction model that accurately predicts CAP-AKI during hospitalization. The web-based prediction platform embedded with the DF model serves as a user-friendly tool for early identification of high-risk patients.https://www.jmir.org/2024/1/e51255
spellingShingle Mengqing Ma
Caimei Chen
Dawei Chen
Hao Zhang
Xia Du
Qing Sun
Li Fan
Huiping Kong
Xueting Chen
Changchun Cao
Xin Wan
A Machine Learning–Based Prediction Model for Acute Kidney Injury in Patients With Community-Acquired Pneumonia: Multicenter Validation Study
Journal of Medical Internet Research
title A Machine Learning–Based Prediction Model for Acute Kidney Injury in Patients With Community-Acquired Pneumonia: Multicenter Validation Study
title_full A Machine Learning–Based Prediction Model for Acute Kidney Injury in Patients With Community-Acquired Pneumonia: Multicenter Validation Study
title_fullStr A Machine Learning–Based Prediction Model for Acute Kidney Injury in Patients With Community-Acquired Pneumonia: Multicenter Validation Study
title_full_unstemmed A Machine Learning–Based Prediction Model for Acute Kidney Injury in Patients With Community-Acquired Pneumonia: Multicenter Validation Study
title_short A Machine Learning–Based Prediction Model for Acute Kidney Injury in Patients With Community-Acquired Pneumonia: Multicenter Validation Study
title_sort machine learning based prediction model for acute kidney injury in patients with community acquired pneumonia multicenter validation study
url https://www.jmir.org/2024/1/e51255
work_keys_str_mv AT mengqingma amachinelearningbasedpredictionmodelforacutekidneyinjuryinpatientswithcommunityacquiredpneumoniamulticentervalidationstudy
AT caimeichen amachinelearningbasedpredictionmodelforacutekidneyinjuryinpatientswithcommunityacquiredpneumoniamulticentervalidationstudy
AT daweichen amachinelearningbasedpredictionmodelforacutekidneyinjuryinpatientswithcommunityacquiredpneumoniamulticentervalidationstudy
AT haozhang amachinelearningbasedpredictionmodelforacutekidneyinjuryinpatientswithcommunityacquiredpneumoniamulticentervalidationstudy
AT xiadu amachinelearningbasedpredictionmodelforacutekidneyinjuryinpatientswithcommunityacquiredpneumoniamulticentervalidationstudy
AT qingsun amachinelearningbasedpredictionmodelforacutekidneyinjuryinpatientswithcommunityacquiredpneumoniamulticentervalidationstudy
AT lifan amachinelearningbasedpredictionmodelforacutekidneyinjuryinpatientswithcommunityacquiredpneumoniamulticentervalidationstudy
AT huipingkong amachinelearningbasedpredictionmodelforacutekidneyinjuryinpatientswithcommunityacquiredpneumoniamulticentervalidationstudy
AT xuetingchen amachinelearningbasedpredictionmodelforacutekidneyinjuryinpatientswithcommunityacquiredpneumoniamulticentervalidationstudy
AT changchuncao amachinelearningbasedpredictionmodelforacutekidneyinjuryinpatientswithcommunityacquiredpneumoniamulticentervalidationstudy
AT xinwan amachinelearningbasedpredictionmodelforacutekidneyinjuryinpatientswithcommunityacquiredpneumoniamulticentervalidationstudy
AT mengqingma machinelearningbasedpredictionmodelforacutekidneyinjuryinpatientswithcommunityacquiredpneumoniamulticentervalidationstudy
AT caimeichen machinelearningbasedpredictionmodelforacutekidneyinjuryinpatientswithcommunityacquiredpneumoniamulticentervalidationstudy
AT daweichen machinelearningbasedpredictionmodelforacutekidneyinjuryinpatientswithcommunityacquiredpneumoniamulticentervalidationstudy
AT haozhang machinelearningbasedpredictionmodelforacutekidneyinjuryinpatientswithcommunityacquiredpneumoniamulticentervalidationstudy
AT xiadu machinelearningbasedpredictionmodelforacutekidneyinjuryinpatientswithcommunityacquiredpneumoniamulticentervalidationstudy
AT qingsun machinelearningbasedpredictionmodelforacutekidneyinjuryinpatientswithcommunityacquiredpneumoniamulticentervalidationstudy
AT lifan machinelearningbasedpredictionmodelforacutekidneyinjuryinpatientswithcommunityacquiredpneumoniamulticentervalidationstudy
AT huipingkong machinelearningbasedpredictionmodelforacutekidneyinjuryinpatientswithcommunityacquiredpneumoniamulticentervalidationstudy
AT xuetingchen machinelearningbasedpredictionmodelforacutekidneyinjuryinpatientswithcommunityacquiredpneumoniamulticentervalidationstudy
AT changchuncao machinelearningbasedpredictionmodelforacutekidneyinjuryinpatientswithcommunityacquiredpneumoniamulticentervalidationstudy
AT xinwan machinelearningbasedpredictionmodelforacutekidneyinjuryinpatientswithcommunityacquiredpneumoniamulticentervalidationstudy