AI-based model predicts who will die from coronavirus | Digital Trends Spanish
Since the first infections appeared, Artificial Intelligence (AI) has played an important role in the fight against COVID-19.
Now, the University of Copenhagen in Denmark has developed an AI-based computational model capable of predicting which patient is most likely to die from the coronavirus.
This computer model can predict the severity of the effects of COVID-19 based on the history of the disease and health data, and was developed by the authors from the first wave of infections.
The results of the study show that artificial intelligence can determine, with a certainty of 90 percent, whether a person who has not yet been infected will die of COVID-19 or not (if they become infected, of course).
At the same time, if she is admitted to hospital, the computer model can predict with 80 percent accuracy whether that person will need a ventilator.
The researchers hope that this advance will help in the management of patients in hospitals, in addition to identifying who are the people who most urgently need a vaccine.
“We started working on the models to help hospitals, as during the first wave they feared they would not have enough ventilators for intensive care patients,” explains Professor Mads Nielsen from the Department of Informatics at the University of Copenhagen.
“Our new findings could also be used to carefully identify who needs a vaccine.”
Thus, the authors are optimistic and hope that their computational model may soon be available in other hospitals in the territory to predict the need for ventilators.