From the Unseen to Foresight: Epidemic Modeling in the Age of Computing and AI

Alessandro Vespignani is Director and Stern Bernberg Family Distinguished Professor at Department of Physics in Northeastern University. His research activity is focused on the study of “techno-social” systems, where infrastructures composed of different technological layers are interoperating within the social component that drives their use and development. Professor Vespignani’s expertise includes contagion models and adaptive behavior, epidemics in structured populations, resilience of coevolving and interdependent networks, conversations in online social networks, global epidemic and mobility model, and mapping world languages through microblogging platforms.

Introducing Epydemix, The ABC of Epidemics

Nicolò Gozzi is Senior Research Scientist at ISI Foundation (Italy), where he contributes to several projects in the field of Computational Epidemiology, including the scientific development of collaborative modeling efforts with the European Centre for Disease Prevention and Control. His work focuses on developing computational models to forecast and project the spread of infectious diseases across multiple scenarios, integrating epidemic and behavioral dynamics, and creating open-source software to support these activities.

Surveillance, Influenza, and Forecasting

Sarabeth Mathis, MPH, is a Health Scientist working for the US Centers for Disease Control and Prevention in the Influenza Division of the National Center for Immunization and Respiratory Diseases.  She primarily works on evaluating and visualizing influenza forecasts for the FluSight influenza forecasting collaboration, modeling influenza disease outcomes mediated by antivirals and vaccines, and analyzing public health datasets. She has experience in infectious and non-infectious disease surveillance.

Matthew Biggerstaff (CDC/NCIRD/ID),
Annabella G Hines, (CDC/NCIRD/ID) (CTR), Rebecca K Borchering (CDC/NCIRD/ID)

Infectious Disease Forecasting with Digital Data Streams

Accurate and timely forecasting of infectious diseases is critical for public health decision-making. Over the past decade, digital data streams such as internet search queries, electronic health records, and pharmacy sales data have emerged as powerful supplements to traditional surveillance systems. In this talk, I will present a synthesis of my research journey in leveraging these data sources for infectious disease prediction, spanning from classical statistical approaches to modern ML architectures. On the statistical side, I will discuss a family of models that combine autoregressive time series structure with penalized regression on Google search data, and their extensions to spatial-temporal modeling, multi-disease settings, and COVID-19 adaptation. On the ML side, I will present recent work on attention-based transformer architectures for time series forecasting, including methods for multivariate dependency modeling, in-context learning, and efficient linear attention. Drawing from our ongoing participation in the CDC FluSight forecasting initiative, I will compare the strengths and limitations of both paradigms and share practical lessons learned from real-time deployment. The talk also offers perspectives on digital data stream as an early signal for infectious disease forecasting.