Innovative Algorithm Predicts COVID-Linked Intensive Care Unit Resource Use
A new algorithm can predict how many patients will need intensive care linked to COVID. This is valuable knowledge when it comes to prioritizing caregivers and ventilators in individual hospitals. The innovation could save lives, according to the UCPH researcher behind the algorithm.
When the COVID-19 pandemic peaked in December 2020, Danish hospitals were under maximum pressure. Hospital staff were running out of steam and the Danish health authority had to make difficult decisions to prioritize treatment. This resulted in particular in 35,500 postponed transactions.
Now, an innovative algorithm will help ease the pressure whenever hospitals face new waves of COVID. Researchers at the University of Copenhagen, among others, have developed the algorithm, which can predict the disease course of COVID patients based on how many of them are most likely or unlikely to require intensive care or ventilation.
This is important for the distribution of staff between hospitals, for example in Denmark, explains one of the authors of the study.
âIf we can see that we will have capacity issues in five days because too many beds are taken at Rigshospitalet, for example, we can better plan and direct patients to hospitals with more space and staff. that such, our algorithm has the potential to save lives, âexplains Stephan Lorenzen, post-doctoral fellow at the Department of Computer Science at the University of Copenhagen.
The algorithm uses individual patient data from Sundhedsplatform (the national health platform), including information on gender, age, medications, BMI, whether or not he smokes, blood pressure and more.
This allows the algorithm to predict how many patients, within one to fifteen days, will need intensive care in the form of, for example, ventilators and constant monitoring by nurses and doctors.
Together with colleagues from the University of Copenhagen, as well as researchers from Rigshospitalet and Bispebjerg Hospital, Lorenzen developed the new algorithm based on health data from 42,526 Danish patients who tested positive for the coronavirus between March 2020 and May 2021.
Predicts the number of ICU patients with 90% accuracy
Traditionally, researchers have used regression models to predict hospital admissions linked to Covid. However, these models did not take into account disease history, age, gender, and other factors.
âOur algorithm is based on more detailed data than other models. This means that we can predict the number of patients who will be admitted to intensive care units or who will need a ventilator within five days with greater than 90% accuracy, âsays Stephan Lorenzen.
In fact, the algorithm provides extremely accurate predictions for the likely number of intensive care patients up to ten days.
âWe make better predictions than comparable models because we are able to more accurately map the potential need for ventilators and 24 hour intensive care for up to ten days. The accuracy decreases slightly beyond that, similar to that of existing algorithmic models used to predict the course of the disease in cases of Covid, “he said.
In principle, the algorithm is ready for deployment in Danish hospitals. As such, the researchers are about to begin discussions with the health professionals concerned.
âWe have shown that data can be used for so many things. And that we in Denmark are fortunate to have so much health information to draw from. Hopefully our new algorithm can help our hospitals avoid Covid overload when a new wave of the disease strikes, âconcludes Stephan Lorenzen.
University of Copenhagen – Faculty of Science
Lorenzen, SS, et al. (2021) Using machine learning to predict intensive care unit resource use during COVID-19 pandemic in Denmark. Scientific reports. doi.org/10.1038/s41598-021-98617-1.