A simple blood test with the help of artificial intelligence can predict who will get a long Covid

New Delhi: As Covid becomes a burden on healthcare providers worldwide, researchers report that a blood test taken during a Covid-19 infection can predict who is most likely to develop prolonged Covid.

The study by a team from University College London in the United Kingdom, published in Lancet eBioMedicine, analyzed proteins in the blood of healthcare workers infected with SARS-CoV-2, comparing them to samples from healthcare workers who were not infected.

Normally, the levels of the proteins in the body are stable, but they found a dramatic difference in the levels of some of the proteins up to six weeks after infection, suggesting disruption of a number of important biological processes.

Using an artificial intelligence (AI) algorithm, they identified a “signature” in the abundance of various proteins that successfully predicted whether the person would continue to report persistent symptoms a year after infection.

If these findings are replicated in a larger, independent group of patients, a test could potentially be offered alongside a polymerase chain reaction (PCR) test that could predict the likelihood that people will develop prolonged Covid.

“The study shows that even mild or asymptomatic Covid-19 disrupts the protein profile in our blood plasma. This means that even mild Covid-19 dramatically affects normal biological processes for at least six weeks after infection,” said lead author Dr Gabi Kaptur.

The tool predicting long Covid still needs to be validated in an independent, larger group of patients.

“However, using our approach, a test that predicts long Covid during initial infection can be introduced quickly and cost-effectively,” Captur added.

More than 40 percent of Covid-19 survivors worldwide, or more than 100 million, have or have had long-term effects after recovery, according to a recent study by researchers at the University of Michigan.

For the UK study, researchers analyzed blood plasma samples from 54 healthcare workers who had PCR or antibody-confirmed infection taken weekly for six weeks in the spring of 2020, comparing them to samples taken during the same period of 102 healthcare workers who were not infected.

The researchers found abnormally high levels of 12 proteins out of 91 tested among those infected with SARS-CoV-2, and that the degree of abnormality tracked with the severity of symptoms.

A machine learning algorithm trained on the participants’ protein profiles was able to distinguish all 11 healthcare workers who reported at least one persistent symptom at one year from infected healthcare workers who did not report persistent symptoms after one year.

“If we can identify people who are likely to develop prolonged Covid, it opens the door to trial treatments such as antivirals at this earlier, initial stage of infection to see if it can reduce the risk of later prolonged Covid said senior author Dr. Wendy Haywood.

Leave a Reply

Your email address will not be published. Required fields are marked *