ORIGINAL RESEARCH article
Front. Pediatr.
Sec. Pediatric Critical Care
Volume 13 - 2025 | doi: 10.3389/fped.2025.1583573
Predictive for Patients with Pneumonia in Pediatric Intensive Care Unit
Provisionally accepted
- 1Middlebury College, Middlebury, Vermont, United States
- 2Stanford University, Stanford, California, United States
- 3Shanghai Literature Institute of Traditional Chinese Medicine, Shanghai, China
- 4Yangpu Hospital of Traditional Chinese Medicine, Shanghai, China
- 5ChengZheng Wisdom (Shanghai) Health Sciences and Technology Co., Ltd, shanghai, China
Select one of your emails
You have multiple emails registered with Frontiers:
Notify me on publication
Please enter your email address:
If you already have an account, please login
You don't have a Frontiers account ? You can register here
Pneumonia is globally recognized as a significant disease burden, particularly among pediatric patients in intensive care units (ICU), where its etiology is complex and the prognosis often poor.This study extracted data from a pediatric-specific intensive care (PIC) database, selecting 795 pediatric pneumonia in the intensive care unit patients from the period of 2010 to 2018. After applying rigorous inclusion and exclusion criteria, 543 cases were confirmed as the study cohort. Through basic statistical analysis of patient's baseline information and 70 laboratory indicators, we identified 25 biomarkers associated with the prognosis of pediatric ICU (PICU) patients. For the construction of the prognostic model, we initially employed a stepwise regression algorithm to filter out 28 prognostically relevant variables, and Spearman and Pearson correlation analyses to identify an intersection of 14 key indicators from the top 20 features. Through parameter tuning and combination across 12 machine learning algorithms, 113 model combinations were formed prediction of patient survival outcomes, with the "Stepglm [both]+GBM" combination achieving the highest average accuracy of 79.4% in both the training and testing sets. Ultimately, this study identified a set of 12 variables, including WBC Count, Glucose, Neutrophils Count, Cystatin C, Temperature(body), Sodium (Whole Blood), Cholesterol (Total), Absolute Lymphocyte Count, Urea, Lactate, and Bilirubin (Total), these variables can provide a dependable basis and novel insights for prognostic evaluation, thereby supporting clinical diagnosis, treatment, and early intervention.
Keywords: Pneumonia, Intensive Care Unit, machine learning algorithms, Paediatrics, predictive models
Received: 26 Feb 2025; Accepted: 02 May 2025.
Copyright: © 2025 Jia, Hu, Ji, Lin, Liu and Wang. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
* Correspondence: Yong Wang, Shanghai Literature Institute of Traditional Chinese Medicine, Shanghai, China
Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.