Artificial intelligence researchers from Johns Hopkins University and Duke University have unveiled a powerful new tool designed to forecast and measure the spread of infectious diseases. The innovation, named PandemicLLM, was introduced in a study published in Nature Computational Science.
PandemicLLM is built on advanced generative language model architecture, similar to the technology behind ChatGPT, and is specifically tailored for epidemic forecasting. The tool analyzes a wide array of variables, including rising infection rates, the emergence of new viral variants, hospitalization data, vaccination coverage, government intervention policies, and demographic characteristics of affected regions.
According to a report by The Independent, the tool is capable of accurately predicting how diseases such as avian influenza, monkeypox, and potentially other future outbreaks could spread. Its core functionality is to simulate different scenarios and provide early warnings by integrating both historical and real-time data.
Lead author Dr. Frank Yang, assistant professor of civil engineering at Johns Hopkins University, explained that the project was born out of lessons learned during the COVID-19 pandemic. “We use both new and old types of information to predict what will happen next,” Yang said. “The global response to COVID-19 showed how critical it is to have proactive tools that can support early decision-making in the face of emerging health crises.”
PandemicLLM represents a major step forward in epidemic preparedness, offering governments, health organizations, and researchers a data-driven, AI-powered approach to anticipating and managing disease outbreaks before they spiral out of control.
