OREANDA-NEWS. An international group of scientists and doctors has found that using artificial intelligence, it is possible to predict the outcome of COVID-19.

The authors of the study, published in the journal Nature Medicine, collected data on 16 thousand patients with COVID-19 from 20 medical institutions around the world. Then, based on the so-called federated testing method, they trained an artificial intelligence model to calculate the required amount of oxygen for ventilated patients and the probability of death one day after hospitalization. The clinical sensitivity of the test was 95 percent and the specificity was 88 percent.

The model has been tested on five continents and has proven its reliability and versatility. Another advantage of the new method is the ability to use it with an unconfirmed diagnosis of COVID-19: the doctor can study the patient's data before the PCR result.

As the spread of coronavirus infection continues, predictive algorithms are important in treating patients, the researchers are convinced.

The results of a new study showed that patients with COVID-19 requiring hospitalization have significantly higher levels of autoimmune antibodies than those with mild disease. The article was published in the journal Nature Communications.