Developing machine learning models for predicting intensive care unit resource use during the COVID-19 pandemic

Lorenzen, S. S.; Nielsen, M.; Jimenez-Solem, E.; Petersen, T. S.; Perner, A.; Thorsen-Meyer, H.-C.; Igel, C.; Sillesen, M.

ABSTRACT Importance: The COVID-19 pandemic has put massive strains on hospitals, and tools to guide hospital planners in resource allocation during the ebbs and flows of the pandemic are urgently needed. Objective: We investigate whether Machine Learning (ML) can be used for predictions of intensive care requirements 5 and 10 days into the future. Design: Retrospective design where health Records from 34,012 SARS-CoV-2 positive patients was extracted. Random Forest (RF) models were trained to predict risk of ICU admission and use of mechanical ventilation after n days (n = 5, 10). Setting: Two Danish regions, encompassing approx. 2.5 million citizens. Participants: All patients from the bi-regional area with a registered positive SARS-CoV-2 test from March 2020 to January 2021. Main outcomes: Prediction of future 5- and 10-day requirements of ICU admission and ventilator use. Mortality was also predicted. Results. Models predicted 5-day risk of ICU admission with an area under the receiver operator characteristic curve (ROC-AUC) of 0.986 and 5-day risk of use of ventilation with an ROC-AUC of 0.995. The corresponding 5-day forecasting models predicted the needed ICU capacity with a coefficient of determination (R2) of 0.930 and use of ventilation with an R2 of 0.934. Performance was comparable but slightly reduced for 10-day forecasting models. Conclusions. Random Forest-based modelling can be used for accurate 5- and 10-day forecasting predictions of ICU resource requirements.