Russian scientists develop algorithms capable of predicting severity of COVID-19

World December 16, 2020, 0:16

The possibility of a lethal outcome is increased depending on such factors as age, male gender, the body mass index over 30 as well as on blood sugar levels

MOSCOW, December 15. /TASS/. Scientists of Russia’s Federal Medical Biological Agency (FMBA) have developed algorithms capable of forecasting with high precision the severity of a course of the coronavirus infection, the agency’s press service told TASS.

"An analysis of a wide range of clinical and diagnostic parameters on a sampling of over 6,900 patients and those who had COVID-19 with varying degree of severity allowed the specialists of Russia’s Federal Medical Biological Agency to create mathematical models (classifiers) of forecasting the severity and the clinical outcome of COVID-19 based on computerized learning algorithms," the statement said.

It is specified that these algorithms with high precision (about 90%) "confirmed a number of clinical factors that allow to forecast the severity of a course of the disease," for example, such indicators as ferritin, C-reactive protein, ESR, neutrophil-lymphocyte ratio, frequency of breathing, and others.

The possibility of a lethal outcome with the coronavirus infection is increased depending on such factors as age, male gender, the body mass index over 30 as well as on blood sugar levels.

"Within the framework of the study, a genome-wide association study (GWAS) was also conducted on a sampling of over 2,000 COVID-19 patients with a symptomless, mild, moderate, and severe forms of the disease. As a result, genetic variants associated with the severity of COVID-19 were detected," the FMBA statement specified. A statistical analysis of the results of the GWAS also allowed to develop a scale of "polygenic risk of development of a severe form of COVID-19." According to the researchers, this diagnostic algorithm allows to forecast the severity of the disease. Currently it is undergoing the validation stage.

"Thus, the pooled data obtained as a result of the study of the significant clinical and diagnostic parameters and human genetic properties allow to build a unified model of evaluation of the joint contribution of both phenotypic and genetic factors, determining the nature of a course of the COVID-19 disease," the FMBA statement said.

Read more on the site →