Washington:
In a new study, a single blood sample from critically ill COVID-19 patients was used for analysis by a machine learning model that uses blood plasma proteins to predict survival.
The research is published in the PLOS Digital Health Journal.
Healthcare systems around the world are struggling to accommodate large numbers of critically ill COVID-19 patients who require special medical attention, especially if they are determined to be at high risk. Clinically established risk assessments in intensive care medicine, such as the SOFA or APACHE II, show only limited reliability in predicting future disease outcomes for COVID-19.
In the new study, researchers examined the levels of 321 proteins in blood samples taken at 349 time points from 50 critically ill COVID-19 patients treated at two independent health centers in Germany and Austria. A machine learning approach was used to find associations between the measured proteins and patient survival.
Fifteen of the patients in the cohort died; the mean time from admission to death was 28 days. For patients who survived, the median length of stay was 63 days. The researchers identified 14 proteins that changed in opposite directions over time for patients who survived compared to patients who did not survive in intensive care.
The team then developed a machine learning model to predict survival based on a single time point measurement of relevant proteins and tested the model on an independent validation cohort of 24 critically ill COVID-10 patients. The model showed high predictive power on this cohort, correctly predicting the outcome for 18 out of 19 patients who survived and 5 out of 5 patients who died (AUROC = 1.0, P = 0.000047).
The researchers concluded that, if validated in larger cohorts, blood protein testing could be useful in both identifying patients at the highest risk of death and testing whether a particular treatment alters an individual patient’s expected trajectory.
(Except for the headline, this story has not been edited by DailyExpertNews staff and has been published from a syndicated feed.)