Synapse
⌘+K
Synapse
PulseExploreJournal ClubResearchersJournals
Instagram
HomeJournal ClubExplore
October 13, 2025Nursing in Critical Care

Machine Learning for Predicting Mortality in Intensive Care Unit Patients: A Prognostic Performance Systematic Review and Meta‐Analysis

View Full Paper
Ask AI
Bookmark
Share

Authors

HSHu SunMKM. M. KangHZHuayu Zhang

Discussion

Loading...

Member takes

Overview

Systematic review evaluates machine learning models predicting ICU mortality, revealing significant predictive performance indicators.

Key Points

  • Machine learning models show a pooled area under the receiver operating characteristic curve of 0.83, indicating strong predictive ability for ICU mortality.
  • The pooled sensitivity and specificity for predicting mortality in ICU patients were found to be 0.72 and 0.81, respectively.
  • This systematic review and meta-analysis synthesized data from 40 studies with 317028 patients to assess machine learning model performance.
  • While showing potential, many studies had limitations due to lack of external validation, calling for multicenter prospective studies.

Cite This Study

Sun et al. (2025) studied this question.

synapsesocial.com/papers/68ec51df42911f61ef8b2148https://doi.org/10.1111/nicc.70206
View Full Paper
Ask AI
Bookmark
Share