Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Identifying a Refractory Shock Phenotype in Pediatric Sepsis Using a Vasoactive-Adjusted Shock Index.

Shock (Augusta, Ga.)·2026
Same author

Association of Albumin Infusion With Differential Response in Pediatric Sepsis and Septic Shock: Retrospective Analysis Using a U.S. Multicenter 2012-2018 Dataset.

Pediatric critical care medicine : a journal of the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies·2026
Same author

A weakly supervised transformer for rare disease diagnosis and subphenotyping from EHRs with pulmonary case studies.

NPJ digital medicine·2026
Same author

Optimizing PICU Resource Management: A Data-Driven Discrete Event Simulation Approach for Capacity and Flow Modeling.

Pediatric critical care medicine : a journal of the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies·2026
Same author

Artificial intelligence-enabled electrocardiogram guidance for pulmonary valve replacement timing in repaired tetralogy of Fallot.

American heart journal·2025
Same author

Large Language Model Symptom Identification From Clinical Text: Multicenter Study.

Journal of medical Internet research·2025
Same journal

Relationship Between Cerebral Oximetry and Outcomes in Post-Cardiac Arrest Patients: A Systematic Review and Meta-Analysis.

Critical care explorations·2026
Same journal

Apnea Testing for Brain Death/Death by Neurologic Criteria in Adults: A Prospective Multicenter Observational Study Assessing the Relationship Between Methodology, Safety, and Duration.

Critical care explorations·2026
Same journal

From Cutaneous Necrosis to Septic Shock: An ICU Case of Pancreatico portal Fistula Complicating Necrotizing Acute Pancreatitis.

Critical care explorations·2026
Same journal

Patient- and Family-Centered Care, Communication, and Relationship Development Along a Technological Continuum in the ICU: A Scoping Review.

Critical care explorations·2026
Same journal

A Qualitative Multicenter Study of Participant Experiences With an App-Based Mindfulness Intervention (Lift 2) for Critical Illness Survivors.

Critical care explorations·2026
Same journal

Using Mixed-Methods Research to Address Stagnant Operative Mortality Rates in Congenital Heart Surgery.

Critical care explorations·2026
See all related articles

Related Experiment Video

Updated: Nov 4, 2025

Comparison of Predictive Performance of Three Lymph Node Staging Systems in Colorectal Signet Ring Cell Carcinoma Based on Machine Learning Model
07:13

Comparison of Predictive Performance of Three Lymph Node Staging Systems in Colorectal Signet Ring Cell Carcinoma Based on Machine Learning Model

Published on: April 18, 2025

305

A Machine Learning Classifier Improves Mortality Prediction Compared With Pediatric Logistic Organ Dysfunction-2

Remi D Prince1,2, Alireza Akhondi-Asl2,3, Nilesh M Mehta2,3

  • 1Tufts University School of Medicine, Boston, MA.

Critical Care Explorations
|May 26, 2021
PubMed
Summary
This summary is machine-generated.

Machine learning models, specifically random forest, significantly outperformed the Pediatric Logistic Organ Dysfunction-2 score in predicting pediatric intensive care unit (PICU) mortality. This advancement offers improved mortality risk estimation for future research, pending external validation.

Keywords:
epidemiologic methodshospital mortalityintensive care unitsmachine learningorgan dysfunction scoresseverity of illness index

More Related Videos

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
07:15

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Published on: August 16, 2020

7.1K
Machine Learning-Based Cough Tone Classification: Diagnostic Exploration of Chronic Obstructive Pulmonary Disease and Respiratory Tract Infections
06:22

Machine Learning-Based Cough Tone Classification: Diagnostic Exploration of Chronic Obstructive Pulmonary Disease and Respiratory Tract Infections

Published on: September 19, 2025

142

Related Experiment Videos

Last Updated: Nov 4, 2025

Comparison of Predictive Performance of Three Lymph Node Staging Systems in Colorectal Signet Ring Cell Carcinoma Based on Machine Learning Model
07:13

Comparison of Predictive Performance of Three Lymph Node Staging Systems in Colorectal Signet Ring Cell Carcinoma Based on Machine Learning Model

Published on: April 18, 2025

305
Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
07:15

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Published on: August 16, 2020

7.1K
Machine Learning-Based Cough Tone Classification: Diagnostic Exploration of Chronic Obstructive Pulmonary Disease and Respiratory Tract Infections
06:22

Machine Learning-Based Cough Tone Classification: Diagnostic Exploration of Chronic Obstructive Pulmonary Disease and Respiratory Tract Infections

Published on: September 19, 2025

142

Area of Science:

  • Biomedical Informatics
  • Machine Learning in Healthcare
  • Pediatric Critical Care Medicine

Background:

  • The Pediatric Logistic Organ Dysfunction-2 (PELOD-2) score is a standard tool for assessing mortality risk in pediatric intensive care units (PICUs).
  • Predictive accuracy of existing scores may be limited, necessitating exploration of advanced analytical methods.
  • Machine learning (ML) offers potential for developing more precise predictive models in critical care settings.

Purpose of the Study:

  • To evaluate the performance of ML algorithms in predicting PICU mortality.
  • To compare the predictive accuracy of ML models against the established PELOD-2 score.
  • To assess the calibration and potential clinical utility of ML-based mortality prediction.

Main Methods:

  • A retrospective study utilizing data from a quaternary care medical-surgical PICU.
  • Patient data from 2013-2017 (n=10,194) were used for training ML models, with data from 2018-2019 (n=4,043) serving as a validation cohort.
  • ML algorithms were trained using the same variables as the PELOD-2 score, and performance was evaluated using AUC-ROC, AUC-PR, and F1 scores.

Main Results:

  • The random forest ML model demonstrated superior performance compared to the PELOD-2 score, achieving a higher area under the receiver operating characteristic curve (0.867 vs. 0.761).
  • The random forest model also showed better area under the precision-recall curve and F1 scores.
  • While the difference diminished after retraining the PELOD-2 model locally, the random forest model maintained superior calibration.

Conclusions:

  • Machine learning models, particularly random forest, can achieve better performance in predicting PICU mortality than traditional logistic regression-based scores like PELOD-2.
  • Enhanced mortality risk prediction can aid in adjusting for illness severity in future research.
  • External validation is recommended before widespread clinical deployment of ML-based prediction models.