Respiratory failure & ventilation
Predicting ICU Mortality in Acute Respiratory Distress Syndrome Patients Using Machine Learning: The Predicting Outcome and STratifiCation of severity in Acute Respiratory Distress Syndrome (ARDS) (POSTCARDS) Study*
Authors' conclusion
Both machine learning and traditional methods lead to promising models to predict ICU death in moderate/severe ARDS patients.
Study context
- Population
- Patients with moderate-to-severe ARDS managed with lung-protective ventilation in multidisciplinary ICUs1,000 development/testing, 303 validation
- Participants
- 1,303 patients
- Comparison
- Machine learning vs logistic regression models
- Outcome
- Discrimination for ICU mortality (development and testing)
Methodological detailsfour prospective, multicenter, observational cohorts, machine learning (genetic algorithm, random forest, extreme gradient boosting)
- Four prospective, multicenter, observational cohorts
- Machine learning (genetic algorithm, random forest, extreme gradient boosting)
Performance
As reported in the abstract
Machine learning performance similar to logistic regression; AUC
0.87
95% CI 0.82–0.91
Further findings
External validation (independent cohort, n = 303)
- Model performance similar to logistic regression model; AUC
- 0.91 (95% CI, 0.87-0.94)
Parts of this Brief are withheld: the automatic checks could not confirm them against the abstract.
Original publication
Predicting ICU Mortality in Acute Respiratory Distress Syndrome Patients Using Machine Learning: The Predicting Outcome and STratifiCation of severity in ARDS (POSTCARDS) Study*. Crit Care Med. Published 31 August 2023. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →