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

The Gradient Energetic Modification Method to Optimize the Balance between Energy and Thermal Stability of Tetrazole-Fused Pyrazolopyrimidines.

Organic letters·2026
Same author

Generalization of negative memories drives the development of psychological distress-related behaviors.

Innovation (Cambridge (Mass.))·2026
Same author

Construction of ultra-heat-resistant explosives <i>via</i> a strategy of building hydrogen-bond networks in heterocyclic bridges.

Chemical communications (Cambridge, England)·2026
Same author

Vitamin D-VDR Signaling Modulates Epithelial-Driven Intra-Alveolar Coagulation via NF-κB in Sepsis-Associated ARDS.

FASEB journal : official publication of the Federation of American Societies for Experimental Biology·2026
Same author

Methylene-Bridged Melt-Castable Explosives with Enhanced Thermal Stability and Detonation Performance.

Organic letters·2026
Same author

Imidazole-1,2,4-Oxadiazole Energetic Materials Functionalized with Carbonyl and Nitroamino Groups: Achieving High Performance and Low Sensitivity.

The Journal of organic chemistry·2026

Related Experiment Video

Updated: Jun 3, 2025

A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
12:18

A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment

Published on: January 11, 2020

7.5K

Machine learning-based 28-day mortality prediction model for elderly neurocritically Ill patients.

Jia Yuan1, Jiong Xiong1, Jinfeng Yang2

  • 1Department of Intensive Care Unit, The Affiliated Hospital of Guizhou Medical University, No. 28, Guiyi Road, Yunyan District, Guiyang, Guizhou 550001, China.

Computer Methods and Programs in Biomedicine
|January 12, 2025
PubMed
Summary

Machine learning accurately predicts 28-day mortality in elderly neurocritical patients. The LightGBM model offers a valuable tool for clinical decision-making and resource allocation in intensive care units (ICUs).

Keywords:
28-day mortalityElderlyMachine learningNeurocriticalPredictive model

More Related Videos

Implementation of a Real-Time Psychosis Risk Detection and Alerting System Based on Electronic Health Records using CogStack
07:31

Implementation of a Real-Time Psychosis Risk Detection and Alerting System Based on Electronic Health Records using CogStack

Published on: May 15, 2020

7.0K
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

6.7K

Related Experiment Videos

Last Updated: Jun 3, 2025

A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
12:18

A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment

Published on: January 11, 2020

7.5K
Implementation of a Real-Time Psychosis Risk Detection and Alerting System Based on Electronic Health Records using CogStack
07:31

Implementation of a Real-Time Psychosis Risk Detection and Alerting System Based on Electronic Health Records using CogStack

Published on: May 15, 2020

7.0K
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

6.7K

Area of Science:

  • Computational medicine and artificial intelligence in critical care.
  • Neurocritical care and patient prognosis.
  • Development and validation of predictive models for patient outcomes.

Background:

  • Increasing elderly population in neurocritical care necessitates improved prognosis prediction.
  • Current tools may not fully capture the complexity of predicting mortality in this demographic.
  • Need for robust machine learning (ML) models to forecast 28-day mortality in intensive care units (ICUs).

Purpose of the Study:

  • To develop and validate machine learning models for predicting 28-day mortality in elderly neurocritically ill patients.
  • To identify key predictors of mortality in this patient population.
  • To assess the generalizability of the developed models using external validation data.

Main Methods:

  • Utilized the Medical Information Mart for Intensive Care IV (MIMIC-IV) database for a cohort of elderly neurocritical patients (ICU stay ≥ 24 h).
  • Employed Lasso regression for feature selection from 58 variables; evaluated seven ML algorithms, including Light Gradient Boosting Machine (LightGBM).
  • Validated the best model using data from Guizhou Medical University Affiliated Hospital and employed SHAP for model interpretability.

Main Results:

  • The LightGBM model achieved high predictive performance (AUC 0.896 internal, 0.812 external validation) in a cohort of 1,773 patients (28.6% mortality).
  • Key mortality predictors identified include partial pressure of arterial carbon dioxide (PaCO2), APACHE II score, white blood cell count, age, and lactate.
  • Kaplan-Meier analysis confirmed the correlation between higher LightGBM scores and decreased survival; model showed consistent performance across subgroups.

Conclusions:

  • The LightGBM model demonstrates significant clinical utility for predicting 28-day mortality risk in elderly neurocritically ill patients.
  • This predictive tool can assist clinicians in optimizing patient management strategies and resource allocation within ICUs.
  • The model's robust performance across diverse patient subgroups highlights its potential for widespread clinical application.