The Biocomplexity Institute (BI) is proud to have supported the US-UK Privacy-Enhancing Technologies (PETs) Prize Challenge by developing a comprehensive synthetic population dataset. This dataset was instrumental for participants in Track B: Pandemic Response and Forecasting, enabling them to create privacy-preserving solutions for pandemic preparedness. 

The PETs Prize Challenge, a collaborative initiative between the United Kingdom and the United States, aimed to drive innovation in privacy-enhancing technologies. Participants were tasked with developing solutions that allow for data-driven insights while upholding individual privacy. Track B specifically focused on enhancing pandemic response capabilities through improved forecasting methods. 

Przemek Porebski, Research Scientist at the Biocomplexity Institute, said: “The data processing, machine learning and simulation infrastructure that we have developed at BI allows us to generate digital similars that represent populations, people activities and social processes at diverse scales, from individuals to global populations. We can incorporate multiple layers of data to create synthetic datasets tailored for different domains, such as epidemiology, agriculture and energy and use simulations to model various scenarios. To enable the PETs Prize Challenge participants to test and refine their models in a controlled, privacy-preserving environment we have leveraged our tools to create a statistically accurate representation of population and a hypothetical disease outbreak – a dataset that does not contain any actual personal information.” 

The synthetic data encompassed two datasets: one representing the state of Virginia (~7.7 million individuals) and the other representing the United Kingdom (~62 million individuals). These datasets were synthesized by combining multiple sources of publicly and commercially available data to produce data that statistically resembles real-world populations. This approach ensured that no entry in the dataset corresponded to any real person's data, maintaining strict privacy standards.

By providing this synthetic population data, BI contributed to the development of innovative, privacy-preserving solutions for pandemic preparedness. These efforts align with the broader goals of the PETs Prize Challenge to harness data-driven insights while preserving citizens' fundamental right to privacy. According to Anil Vullikanti, Professor at the University of Virginia, "US-UK collaborations like this PET challenge are vital in accelerating the development of privacy-enhancing technologies that can make a meaningful impact on public health initiatives. The synthetic datasets provided by the Biocomplexity Institute have been an invaluable tool for modeling pandemic response strategies that respect privacy while leveraging data for better forecasting and preparedness." 

The challenge culminated in March 2023, with winners announced at the Summit for Democracy. The innovative solutions presented have the potential to transform pandemic preparedness and response, ensuring that data privacy is maintained even in critical public health initiatives. More information about these datasets is available here. 

The Biocomplexity Institute is dedicated to solving complex problems that cannot be solved within the narrow confines of a single discipline or by a single researcher – we have developed a unique transdisciplinary team science approach to tackle such problems. We work to foster global collaborative networks to advance the research agenda and develop innovative experiential learning programs for students.