Clinical characteristics of hospitalized children with community-acquired pneumonia and respiratory infections: Using machine learning approaches to support pathogen prediction at admission

Tu-Hsuan Chang; Yun-Chung Liu; Siang-Rong Lin; Pei-Hsin Chiu; Chia-Ching Chou; Luan-Yin Chang; Fei-Pei Lai

doi:10.1016/j.jmii.2023.04.011

Clinical characteristics of hospitalized children with community-acquired pneumonia and respiratory infections: Using machine learning approaches to support pathogen prediction at admission

J Microbiol Immunol Infect. 2023 Aug;56(4):772-781. doi: 10.1016/j.jmii.2023.04.011. Epub 2023 May 1.

Authors

Tu-Hsuan Chang¹, Yun-Chung Liu², Siang-Rong Lin³, Pei-Hsin Chiu³, Chia-Ching Chou⁴, Luan-Yin Chang⁵, Fei-Pei Lai⁶

Affiliations

¹ Department of Pediatrics, Chi Mei Medical Center, Tainan City, Taiwan.
² Department of Pediatrics, National Taiwan University Hospital, College of Medicine, National Taiwan University, Taipei City, Taiwan.
³ Institute of Applied Mechanics, National Taiwan University, Taipei City, Taiwan.
⁴ Institute of Applied Mechanics, National Taiwan University, Taipei City, Taiwan. Electronic address: ccchou@iam.ntu.edu.tw.
⁵ Department of Pediatrics, National Taiwan University Hospital, College of Medicine, National Taiwan University, Taipei City, Taiwan. Electronic address: lychang@ntu.edu.tw.
⁶ Graduate Institute of Biomedical Electronics and Bioinformatics, Taipei City, National Taiwan University, Taiwan; Department of Computer Science and Information Engineering, National Taiwan University, Taipei City, Taiwan; Department of Electrical Engineering, National Taiwan University, Taipei City, Taiwan.

PMID: 37246060
DOI: 10.1016/j.jmii.2023.04.011

Abstract

Background: Acute respiratory infections (ARIs) are common in children. We developed machine learning models to predict pediatric ARI pathogens at admission.

Methods: We included hospitalized children with respiratory infections between 2010 and 2018. Clinical features were collected within 24 h of admission to construct models. The outcome of interest was the prediction of 6 common respiratory pathogens, including adenovirus, influenza virus types A and B, parainfluenza virus (PIV), respiratory syncytial virus (RSV), and Mycoplasma pneumoniae (MP). Model performance was estimated using area under the receiver operating characteristic curve (AUROC). Feature importance was measured using Shapley Additive exPlanation (SHAP) values.

Results: A total of 12,694 admissions were included. Models trained with 9 features (age, event pattern, fever, C-reactive protein, white blood cell count, platelet count, lymphocyte ratio, peak temperature, peak heart rate) achieved the best performance (AUROC: MP 0.87, 95% CI 0.83-0.90; RSV 0.84, 95% CI 0.82-0.86; adenovirus 0.81, 95% CI 0.77-0.84; influenza A 0.77, 95% CI 0.73-0.80; influenza B 0.70, 95% CI 0.65-0.75; PIV 0.73, 95% CI 0.69-0.77). Age was the most important feature to predict MP, RSV and PIV infections. Event patterns were useful for influenza virus prediction, and C-reactive protein had the highest SHAP value for adenovirus infections.

Conclusion: We demonstrate how artificial intelligence can assist clinicians identify potential pathogens associated with pediatric ARIs upon admission. Our models provide explainable results that could help optimize the use of diagnostic testing. Integrating our models into clinical workflows may lead to improved patient outcomes and reduce unnecessary medical costs.

Keywords: Children; Community-acquired pneumonia; Machine learning; Pathogens prediction; Respiratory infections.

MeSH terms

Adenoviridae
Adenoviridae Infections*
Artificial Intelligence
C-Reactive Protein
Child
Child, Hospitalized
Humans
Infant
Influenza, Human*
Machine Learning
Mycoplasma pneumoniae
Parainfluenza Virus 1, Human
Pneumonia*
Respiratory Syncytial Virus, Human*
Respiratory Tract Infections* / diagnosis

Substances

C-Reactive Protein