The global adoption of the Internet, mobile phones and related technologies has resulted in an unprecedented amount of digital data that have been used in studying health and disease. Specifically, digital data have been used to study how infectious diseases spread, sentiments towards public health interventions (such as, vaccines), risk factors for chronic conditions (e.g., smoking, alcohol abuse and unhealthy diets), and how public health information, and misinformation diffuses through social networks. Organizations such as the U.S. Centers for Disease Control and Prevention (CDC) and the World Health Organization (WHO) have used data from digital sources for the identification of disease outbreaks, and the 2005 International Health Regulations recognized event-based informal surveillance as an important source for epidemic intelligence.

Additionally, advances in machine learning and artificial intelligence (AI) are creating unique opportunities for developing new solutions and tools to improve health globally. Applications of machine learning and AI have enabled rapid processing of large amounts of data from digital and non-digital sources to identify previously undiscovered patterns of
health and disease, forecast disease trends and understand how inequalities in the built environment impacts health outcomes. Given the rapidly growing interest in applying machine learning and AI to public health, it is imperative for researchers and practitioners to gather and discuss advances, opportunities and challenges.

The AI & Health track aims to bring together public health practitioners and researchers working on data mining, crowdsourcing, social media, and AI applications for relevant discussions on the opportunities and ethics on the use of AI in public health. Contributed and invited talks will focus on novel applications of AI/machine learning to digital and non-digital data to improve health and control disease in communities. We will also aim to have representation of speakers from low, and middle income countries.