Related Experiment Video

Updated: Jun 20, 2026

Repeated Measurement of Respiratory Muscle Activity and Ventilation in Mouse Models of Neuromuscular Disease
09:24

Repeated Measurement of Respiratory Muscle Activity and Ventilation in Mouse Models of Neuromuscular Disease

Published on: April 17, 2017

Comparative machine learning modeling of resting energy expenditure estimation in mechanically ventilated children

Chenyu Li1, Zhiyuan Zhu1, Shilin Wang1

  • 1Department of Pediatric Intensive Care Unit, National Center for Cardiovascular Disease and Fuwai Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing, China.

European Journal of Clinical Nutrition
|June 18, 2026
PubMed

Insights

Machine learning models accurately predict resting energy expenditure (REE) in children post-cardiac surgery, outperforming traditional methods. This aids in optimizing nutritional support for better recovery outcomes.

Area of Science:

  • Pediatric critical care medicine
  • Biomedical engineering
  • Nutritional science

Background:

  • Optimal nutrition is crucial for pediatric recovery after cardiac surgery.
  • Metabolic demands exhibit significant inter-individual variability.
  • Accurate assessment of resting energy expenditure (REE) is essential for tailored nutritional plans.

Purpose of the Study:

  • To develop and evaluate machine learning (ML) models for predicting postoperative REE in children undergoing cardiac surgery.
  • To identify key factors influencing REE in this patient population.
  • To compare the performance of ML models against conventional predictive equations.

Main Methods:

  • Retrospective analysis of 278 children who underwent indirect calorimetry (IC) post-cardiac surgery.
  • Training nine ML algorithms to predict REE using clinical and demographic data.
  • Assessing model agreement with measured REE using Bland-Altman analysis and SHAP for interpretability.

Main Results:

  • ML models demonstrated competitive accuracy, with the best models achieving an R-squared of 0.64.
  • The root mean square error (RMSE) for ML models ranged from 226 to 281 kcal/d, comparable to conventional equations.
  • Key predictors of REE included weight, age, height, preoperative albumin, CK-MB, vasoactive inotropic score, CPB time, LVEDD Z-score, gender, and NT-proBNP.

Conclusions:

  • Developed ML models show promise in predicting REE in pediatric cardiac surgery patients.
  • Certain ML models outperformed traditional equations in estimating postoperative REE.
  • These findings support the use of ML for personalized nutritional management, improving metabolic response to surgical stress.
Abstract

Related Concept Videos

Mechanical Ventilation III: Noninvasive Ventilation01:23

Mechanical Ventilation III: Noninvasive Ventilation

Noninvasive positive-pressure ventilation (NIPPV), continuous positive airway pressure (CPAP), and bilevel positive airway pressure (BiPAP) are essential methods in respiratory care. These ventilation techniques offer unique benefits for patients with various respiratory conditions, providing adequate support without requiring intubation. Let's explore how each method is crucial in improving patient outcomes and enhancing respiratory therapy.
Noninvasive Positive-Pressure Ventilation (NIPPV)
Mechanical Ventilation I: Indication and Settings01:29

Mechanical Ventilation I: Indication and Settings

Mechanical ventilation is a life-saving technique for managing acute respiratory failure and other respiratory complications. The process involves using a machine known as a ventilator to supply oxygen to the lungs and assist in removing carbon dioxide. It serves as a bridge to long-term mechanical ventilation or a temporary measure until ventilatory support is discontinued. The ventilator can maintain this function for a prolonged period, providing critical support for patients until they can...