News – Using patient data, artificial intelligence can make a 90 percent accurate assessment of whether or not a person will die from COVID-19, according to a new study at the University of Copenhagen. Body mass index (BMI), sex and high blood pressure are among the most important factors. The research can be used to predict the number of patients in hospitals, who need respiration and decide who should be vaccinated first.
Artificial intelligence can predict the most likely death from the coronavirus. By doing this, it can also help decide who should be in front of the line for the valuable vaccines that are now being delivered across Denmark.
The results are from a newly published study by researchers at the Department of Computer Science at the University of Copenhagen. Since the first wave of COVID epidemic, researchers have been working to develop computer models that can predict, based on disease history and health data, the severity of COVID-19 infection. influencing people.
Based on patient data from the Danish Capital Region and the Zealand Region, the results of the study show that artificial intelligence, with up to 90 percent certainty, can determine the death of an unprotected person who has not yet been infected with COVID-19 or do not die if they are unfortunate enough to be infected. Once admitted to the hospital with COVID-19, the computer can predict with 80 percent certainty whether the person will need to breathe.
“We started working on the models to help hospitals, because in the first wave, they feared that they did not have enough respiration for patients with intensive care. Our new findings could be used to find out who needs vaccinations, “explained Professor Mads Nielsen of the University of Copenhagen ‘s Department of Computer Science.
Older men with high blood pressure are at highest risk
The researchers fed a computer program with health data from 3,944 Danish COVID-19 patients. This trained the computer to identify patterns and correlations in patients’ prior illnesses and in their effects against COVID-19.
“Our results confirm, not surprisingly, that age and BMI are the most accurate parameters of how badly a person is affected by COVID-19. But it is likely to die. or that he may also be able to breathe if you are male, have high blood pressure or a brain disease, “Mads Nielsen explains.
The diseases and health factors that, according to the study, have the greatest impact on whether a patient discontinues relief after being ingested with COVID-19 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 have found that it may be sensible to move them up in the vaccine queue, to avoid any risk of entry. and finally getting some relief, ”says Nielsen.
Predicting respiratory needs is essential
Researchers are currently working with the Danish Capital Department to exploit this new batch of products in practice. They hope that artificial intelligence will soon be able to help the nation ‘s hospitals by continuously predicting the need for respirators.
“We are working towards a goal that we should be able to predict the need for inhalers five days ahead by giving the computer access to health data on everything advanced COVID in the sector, “said Mads Nielsen, adding:
“The computer will never be able to replace a doctor’s assessment, but it can help doctors and hospitals see many patients with COVID-19 infection at once and set ongoing priorities.”
However, technical work is still pending the availability of health data from the department for the computer and beyond to quantify the risk to the infected patients. The research was conducted in collaboration with Rigshospitalet and Bispebjerg and Frederiksberg Hospital.
###
Information:
- Data is processed on Computerome, a secure supercomputer for personal data, and under the license of the Danish Patient Safety Authority, data owners and other relevant authorities.
- False information predicts with 90 percent accuracy the death of an infectious patient of COVID-19.
- Once a person is hospitalized with COVID-19, artificial intelligence can predict whether the person should breathe with 80 percent accuracy.
- BMI, age, high blood pressure, being male, meningitis, COPD, asthma, diabetes and heart disease are the factors that increase artificial weight with the risk of getting pregnant. into the breath.
- The computer models are based on health data from 3,944 COVID-19 patients from the Department of Capital and the Department of Zealand.
- The article is published in the scientific journal Scientific Reports.
- The study is supported by the Novo Nordisk Foundation and the Innovation Fund.