Artificial intelligence is everywhere, and now a group of developers has created AI software that can tell whether you are likely to die from Covid-19 using health data.
The researchers at the University of Copenhagen fed a computer program with health data from 3,944 Danish patients with COVID-19, as well as any underlying conditions.
They then trained him to look for patterns in a patient’s previous illness to determine the risk factors and potential outcome of Covid-19 and found that BMI, age and male gender were the highest risk factors when it came to probability of dying.
The results show that AI can, with up to 90 percent certainty, determine whether an uninfected person will die of the disease if they are unlucky enough to catch it.
The results of the new tool can help health officials determine who should be at the front of the queue for a limited supply of vaccines, said lead author Mads Nielsen.
Artificial intelligence is everywhere, and now a group of developers has created AI software that can tell if you are likely to die from Covid-19 using health data (stock image)
Once admitted to the hospital with Covid-19, the computer’s software can predict with 80 percent accuracy whether a person will need a respirator, the team found.
Certain diseases and health factors have a greater influence on whether a patient ends up using a respirator than others after being infected, according to the study.
In order of priority are: BMI, age, hypertension, being male, neurological diseases, COPD, asthma, diabetes and heart disease.
The group most at risk of dying from coronavirus if they contract the disease are white, fat, elderly men with hypertension, according to the researchers.
For diagnosed patients, age and BMI were among the most relevant characteristics for predicting hospital admission and ventilatory treatment, they found.
Hypertension – high blood pressure – was the most important resource for predicting admission to the ICU and, in fact, an important resource for all models.
For patients who had to be hospitalized, the most relevant factors for disease progression were age, BMI, hypertension and the presence of dementia.
“We started to work on the models of hospital care, because in the first wave they feared they would not have enough respirators for patients in intensive care,” explained Professor Nielsen.
“Our results demonstrate, unsurprisingly, that age and BMI are the most decisive parameters for how severely a person will be affected by COVID-19,” he said.
“But the likelihood of dying or ending up on a respirator also increases if you’re a man, have high blood pressure or a neurological disease,” explains Mads Nielsen.
The AI was trained to look for patterns in a patient’s previous illness to determine risk factors and the potential outcome of Covid-19 and found that BMI, age and being a man were the highest risk factors when it came to the likelihood of dying . Stock Image
For hospitalized patients requiring ICU admission compared to hospitalized patients without ICU admission, only male gender, BMI, dementia and hypertension differed between patients.
Patients admitted to the ICU were, moreover, more likely to be smokers, older and male, the team found.
Those who died of the disease were also more likely to suffer from hypertension, diabetes, heart disease, heart failure, arrhythmias, stroke, COPD or asthma, osteoporosis, dementia, mental disorders, neurological diseases, cancer, chronic kidney failure and use of dialysis.
“For those affected by one or more of these parameters, we found that it might make sense to move them up the line in the vaccine, to avoid any risk of getting flexed and ending up on a respirator,” says Nielsen.
The team behind the study is now working to update its model with new data from the latest coronavirus wave in Denmark.
They hope that artificial intelligence will soon be able to help hospitals in the country, continuously predicting the need for respirators.
“We are working to predict the need for respirators five days in advance, giving the computer access to the health data of all positive COVIDs in the region,” said Mads Nielsen.
“The computer will never be able to replace a doctor’s assessment, but it can help doctors and hospitals see many patients infected with COVID-19 at once and set ongoing priorities.”
The results were published in the journal Scientific Reports.