Researchers develop first AI tool to detect COVID-19 probability

A team of researchers from the University of Vermont and Cedars-Sinai has developed the first software to predict the probability of COVID-19 infections.

AI-COVID uses details from routine blood tests and can assign the probability of a patient being COVID-19 negative with high accuracy. This can help hospitals reduce the number of patients referred for PCR testing.

According to Republic, the study has been published in the Journal of Medical Internet Research. The lead author and the University of Vermont Assistant Professor Timothy Plante noted how experts have “better understanding” of how to move forward with the care of COVID-19 patients. The current standard diagnostic for the disease is PCR testing that requires specific sampling such as nasal swab along with well-equipped laboratory.