COVID-variant waves could be predicted using a new AI model.
After two weeks, the model’s accuracy surpassed 80% in predicting roughly 73% of variants in each country that will cause at least 1,000 instances per 10 lakh people within three months.
A new artificial Intelligence (AI) model that can predict the formation of COVID-variant waves has been revealed by a recent study. An AI algorithm was used in the study, which was carried out by a group from the Massachusetts Institute of Technology in the United States and the Hebrew University-Hadassah Medical School in Israel, to examine 9 million genetic sequences of the SARS-CoV-2 virus from 30 different nations.
After two weeks, the model’s accuracy surpassed 80% in predicting roughly 73% of variants in each country that will cause at least 1,000 instances per 10 lakh people within three months.
Drawing attention to the shortcomings of existing models in terms of forecasting variant-specific spread, the group put up a machine learning-based risk assessment approach. This method improves early identification and forecasts the future spread of recently discovered variations by utilizing genomic data unique to variants as well as epidemiological data.
In order to forecast the future trajectory of diverse infectious diseases, the researchers proposed the possible expansion of this modeling approach to additional respiratory viruses, such as influenza, avian flu viruses, or other coronaviruses.
According to the study’s authors, this work improves early signals and forecasts the future dissemination of recently discovered variations by combining variant-specific genomic data with epidemiological data.
They claimed that the innovative modeling technique might be used to forecast the future course of other infectious diseases as well as other respiratory viruses including influenza, avian flu viruses, or other coronaviruses. Despite shifting waves and variants during the research period, the model remained accurate and performed consistently across racial, ethnic, and gender categories. It is applicable to other dynamic diseases as well.
When COVID-19 initially appeared, medical professionals struggled to comprehend patients’ illnesses in the face of quickly evolving research and symptoms. The announcement said that while predictive models were created during the epidemic, the majority of them do not take into consideration changes in patient characteristics and outcomes that occur in real time.
The most recent AI tool continuously assesses the predicted performance and updates itself automatically upon identifying deviations. The researchers contend that in order to give physicians reliable 28-day survival estimates, this is essential.
