The computer model can determine if you die of COVID-19

Artificial intelligence can predict who is likely to die of the coronavirus. Doing so can also help decide who should be at the forefront of the costly vaccines currently being administered in Denmark.

The result is from a newly published study by researchers at the Department of Computer Science at the University of Copenhagen. Since the first wave of the COVID-19 pandemic, researchers have been working to develop computer models that can predict how badly people will be affected by COVID-19 based on their medical history and health data.

Based on patient data from the capital region of Denmark and the region of Zealand, the results of the study show that artificial intelligence can determine with up to 90 percent certainty whether an uninfected person who is not yet infected with COVID-19 will die or not if they are unlucky. enough to become infected. After being admitted to the hospital with COVID-19, the computer can predict with 80 percent accuracy whether the person needs a respirator.

“We started working on the models to help hospitals because they were afraid they did not have enough respirators for intensive care patients. Our new findings could also be used to accurately identify who needs a vaccine,” he explains. Professor Mads. Nielsen from the Department of Computer Science at the University of Copenhagen.

Older men with high blood pressure are at greatest risk

The researchers conducted a computer program with health data from 3,944 Danish COVID-19 patients. It trained the computer to recognize patterns and correlations in patients’ previous illnesses and in their effects against COVID-19.

“Our results show, surprisingly, that age and BMI are the most crucial parameters for how severely a person will be affected by COVID-19. But the probability of him dying or ending up in a respirator is also increased if you are male. , blood pressure or a neurological disease, ‘explains Mads Nielsen.

The diseases and health factors that, according to the study, have the greatest influence on whether a patient ends up in a respirator after being infected with COVID-19 are in order of priority: BMI, age, high blood pressure, being male, neurological diseases, COPD, asthma, diabetes and heart disease.

“For those affected by one or more of these parameters, we found that it may make sense to move it up in the vaccine line, to avoid the risk of them bending and eventually ending up in a respirator,” says Nielsen.

Predicting breathing needs is a must

Researchers are currently working with the capital region of Denmark to put this fresh number of results into practice. They hope that artificial intelligence will soon be able to help the country’s hospitals by constantly predicting the need for respirators.

“We are working towards a goal that we should be able to predict the need for respirators five days in advance by giving the computer access to health data on all COVID positives in the region,” says Mads Nielsen, adding:

“The computer will never be able to replace a physician’s assessment, but it can help physicians and hospitals see many COVID-19 infected patients at once and set ongoing priorities.”

However, technical work is still being done to make health data from the region available to the computer and thereafter to calculate the risk for the infected patients. The research was carried out in collaboration with the Rigshospitalet and the Bispebjerg and the Frederiksberg Hospital.