The team used the patterns that emerged from this analysis to create a machine-learning-enabled risk assessment model, according to findings that were published in the journal PNAS Nexus. It is claimed that this will identify 72.8% of the variants in each nation that will result in at least 1000 instances per million people over the course of the next three months.
In India, the number of COVID-19 cases is progressively increasing once more. Amidst this, researchers at the US’s Massachusetts Institute of Technology (MIT) have created an artificial intelligence (AI) model that can identify SARS-CoV-2 mutations that could lead to fresh infection waves. The medical industry is one of the major applications of artificial intelligence, and the models that are currently in use to forecast viral transmission are unable to anticipate the variant-specific spread of infection.
Retsef Levi of MIT’s Sloan School of Management led a team that examined the factors that could contribute to the viral transmission using a genomic sequence analysis of nine million SARS-CoV-2 sequences. These sequences were gathered from 30 different nations by the Global Initiative for Sharing Avian Influenza Data (GISAID). The data also contains vaccination rates and illness rates.
The team used the patterns that emerged from this analysis to create a machine-learning-enabled risk assessment model, according to findings that were published in the journal PNAS Nexus. It is claimed that this will identify 72.8% of the variants in each nation that will result in at least 1000 instances per million people over the course of the next three months.
The researchers said, “This work provides an analytics framework that leverages multiple data sources, including genetic sequences data and epidemiological data via machine-learning models to provide improved early signals on the spread risk of new SARS-CoV-2 variants”.
After one week of observation following detection, the predicted accuracy is 72.8%; however, after two weeks of observation, the performance rises to 80.1%.
More research in this area is planned, and scientists believe that additional respiratory viruses including influenza, avian flu viruses, or other coronaviruses may be treated using a similar strategy.
