Anesthesiology
-
Multicenter Study
Predicting Intensive Care Delirium with Machine Learning: Model Development and External Validation.
Delirium poses significant risks to patients, but countermeasures can be taken to mitigate negative outcomes. Accurately forecasting delirium in intensive care unit (ICU) patients could guide proactive intervention. Our primary objective was to predict ICU delirium by applying machine learning to clinical and physiologic data routinely collected in electronic health records. ⋯ Machine learning models trained with routinely collected electronic health record data accurately predict ICU delirium, supporting dynamic time-sensitive forecasting.
-
Monitoring and controlling lung stress and diaphragm effort has been hypothesized to limit lung injury and diaphragm injury. The occluded inspiratory airway pressure (Pocc) and the airway occlusion pressure at 100 ms (P0.1) have been used as noninvasive methods to assess lung stress and respiratory muscle effort, but comparative performance of these measures and their correlation to diaphragm effort is unknown. The authors hypothesized that Pocc and P0.1 correlate with diaphragm effort and lung stress and would have strong discriminative performance in identifying extremes of lung stress and diaphragm effort. ⋯ Pocc and P0.1 correlate with lung stress and diaphragm effort in the preceding hour. Diagnostic performance of Pocc and P0.1 to detect extremes in these parameters is reasonable to excellent. Pocc is more accurate in detecting high diaphragm effort.