AI tool is built to detect which COVID patients will recover from the disease based of blood protein levels
Researchers may have developed a new tool that uses machine learning to better predict health outcomes for hospitalized Covid patients, and help physicians make more informed treatment decisions.
A German research team from Charity-University Medicine in Berlin – one of the country’s largest university hospitals – developed an Artificial Intelligence tool that can estimate how well an infected person will fare based off of a blood sample.
The levels of fourteen proteins found in a person’s blood can indicate whether a person who suffers a severe enough hospitalization will survive or die from the virus, and the tool developed by researchers can accurately asses their risk.
In times of crisis, where resources are especially scarce, the tool can help determine what patients require the most intensive care to survive, and who is more fit to fight off the virus themselves.
Using blood samples from Covid patients, a German research team has found that levels of 14 proteins can help determine whether a person survives the virus. They built a machine learning tool that accurately predicted the outcome of 23 of 24 patients
The tool is simple to use and can take tough decisions out of the hands of doctors and instead into a more accurate AI system. They hope it can assist health care systems in times of need. (file photo)
‘Our study shows that a combination of markers, combined in a risk prediction model based on artificial intelligence, can fairly well predict the probability that an individual patient will die or survive,’ Dr Florian Kurth, a researcher from Charity-University and co-author of the study, said in a press release.
Researchers, who published their findings Tuesday in PLOS Digital Health, first gathered data from 50 critically ill Covid patients from Germany and Austria for the study.
Blood samples from the patients were gathered, and researchers used these samples to measure for proteins and other biomarkers.
Eventually, 15 of the 50 patients ended up dying. Researchers searched for trends in protein levels among those that survived and those that did not.
Using their findings from the original 50 patients, they built an artificial intelligence system that could predict whether a person will die from Covid based on these indicators.
That system was then trialed in a group of 24 real world Covid patients who were receiving treatment at their hospital.
Of the study group, 19 patients ended up surviving their illness and five succumbed to it. The machine learning tool predicted all five deaths, and correctly identified that 18 of 19 patients would survive.
‘We found 14 proteins which over time changed in opposite directions for patients who survive compared to patients who do not survive in intensive care,’ Kurth said.
‘Interestingly, the plasma levels of all of those proteins had been found to be altered by COVID-19 before [and] makes us particularly confident in our findings.’
Researchers note that their study included a very small sample size, with the machine learning only using samples from 50 patients, and only 24 patients included in its trialing.
Early results are promising, though, and they hope to get the opportunity to trial it with a larger population to determine whether they have actually developed a system that can be crucial to treatment decisions going forward.
Overwhelmed hospitals with overworked staff and not enough resources to go around has been a universally recognized scene during the pandemic.
With every wave of the pandemic that comes, basically everywhere hospitals once again get filled to the brim with Covid patients.
Even during the most recent Omicron wave, a strain that is generally more mild than previous versions of the virus, there are reports of hospitals around the country having trouble dealing with a surge of patients.
The situation has even forced some hospitals into having to ration care with doctors having to make tough decisions when there just are not enough resources to go around.
Systems like the one developed by the German team can help physicians make more efficient use of limited resources, and also reduce some of the emotional toll of these decisions since they are made – accurately – by an intelligence system rather than just the doctor themselves making the choice.