Respiratory failure & ventilation
Development of a machine learning model for prediction of the duration of unassisted spontaneous breathing in patients during prolonged weaning from mechanical ventilation
Authors' conclusion
The developed machine learning model showed informed results when predicting the spontaneous breathing capacity of a patient in prolonged weaning, however lacking prognostic quality required for direct translation to clinical use.
Study context
- Population
- Patients in prolonged weaning from mechanical ventilation in a specialised weaning unit18.948 patient-days
- Participants
- 1,018 patients
- Outcome
- Prediction of the duration of unassisted spontaneous breathing (spontaneous breathing trials (SBTs)) on the next day
Methodological detailsmachine learning, histogram-based gradient boosting, separate training and evaluation data sets, single specialised weaning unit
- Machine learning
- Histogram-based gradient boosting
- Separate training and evaluation data sets
- Single specialised weaning unit
Main finding
As reported in the abstract
Model performance
Classifier (increase of duration): ROC-AUC
0.713
Regressor, absolute duration: mean absolute error
2:50 h
Regressor, day-to-day difference: mean absolute error
2:47 h
Original publication
Development of a machine learning model for prediction of the duration of unassisted spontaneous breathing in patients during prolonged weaning from mechanical ventilation. J Crit Care. Published 26 March 2024. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →