Artificial intelligence for rapid epidemic analysis

Early detection of infectious disease outbreaks can prevent small clusters of cases from becoming costly epidemics. However, many under-resourced health systems lack adequate capacity for rapid laboratory diagnosis, especially for rare and emerging diseases. Instead, many outbreaks are only identified based on symptoms and only fully verified after many months. Furthermore, analytical tools to track and forecast outbreaks are rarely fit for use in low-resource settings. I will develop analytical methods and tools to help identify outbreak causes and respond to outbreaks more quickly and effectively, including by combining public health and data science approaches to rapidly predict the causes and properties of infectious disease outbreaks. Over the course of this fellowship, I will develop these methods, validate them against historical outbreak data, and finally build and field test software tools to be used by public health officials, empowering them to control outbreaks locally without international outbreak responses.