Artificial intelligence predicts who will die from Corona patients in intensive care

With the rapid spread of the omicron mutant around the world, hospitals everywhere are prioritizing the treatment of critically ill coronavirus patients or those at risk of life-threatening complications.

In the intensive care unit, it is often difficult to determine who may survive the infection, and who may end up succumbing to the disease.

That could change very soon thanks to a new AI tool that can predict the survival outcome for severe cases of COVID-19 weeks in advance, all from a single blood sample.

The new model, described in paper Published in PLOS Digital Health on Tuesday, it helps clinicians develop more informed treatment plans for COVID-19 patients during initial hospitalization, to reduce the odds of mortality and improve patient care.

“Clinical forms of COVID-19 are exceptionally diverse, ranging from asymptomatic infection to severe illness and death,” Florian Court, a clinical researcher at Charity-University Medicine in Berlin and lead author of the new study, said in a statement. It is difficult to estimate the individual risk of a patient’s condition from deterioration to death due to COVID-19.. The new AI model can very well predict the probability of a patient dying or surviving COVID-19.”

The new study was conducted in two phases. First, the researchers studied hundreds of blood samples from 50 critically ill COVID-19 patients treated in Germany and Austria, to see how levels of 321 different proteins changed during the course of infection. All patients were in intensive care, requiring ventilation and additional treatment, 15 patients died and 35 survived.

The new artificial intelligence tool can predict the survival outcome of severe cases of Corona (Al Jazeera)

The researchers learned that 14 proteins were most associated with either COVID-19 survival or death, and that these protein levels were altered early by disease, the most important of which were proteins involved in blood clotting and antibody function.

In the second part, the new AI model is built and trained to diagnose COVID-19 patients based on the levels of these 14 proteins in a single blood sample.

Kurth and colleagues tested this model on samples from a new cohort of 24 critically ill patients. Of this group, 19 patients survived and 5 died, and the model was able to accurately predict outcomes for all but one of these patients. This prognosis was “much better” than those used in current clinical care risk assessments, Kurt said.

The current study is based on a very small sample size, so the authors want to run the model through much broader tests to prove that it can be a valuable predictor of COVID-19 outcomes in real-world hospital settings.

That will take a long time, but they also hope that the research findings will help other researchers better understand the mechanisms behind COVID-19, and the types of treatments that should be developed to halt the disease’s progression in general.

“There is hope that the studies here will open the doors to a new future of medicine, where we can early predict a disease based on a blood test alone,” Kurth says.

Leave a Reply

Your email address will not be published.