Collaborative Research: NSF-CSIRO: HCC: Small: NSF-CSIRO: Understanding Bias in AI Models for the Prediction of Infectious Disease Spread

Collaborative Research: NSF-CSIRO: HCC: Small: NSF-CSIRO: Understanding Bias in AI Models for the Prediction of Infectious Disease Spread Collaborative Research: NSF-CSIRO: HCC: Small: NSF-CSIRO: Understanding Bias in AI Models for the Prediction of Infectious Disease Spread Infectious disease spread within the human population can be conceptualized as a complex system composed of individuals that interact and transmit viruses via spatiotemporal processes that manifest across and between scales. The complexity of this system ultimately means that infectious disease spread is difficult to understand, predict, and effectively respond to. Machine learning and artificial intelligence (AI) have achieved impressive results in other learning and prediction tasks, but have not yet been widely applied to epidemic prediction and preparedness. This lack of pervasion of AI methods in this domain may be contributed to the lack of explainability of common ``black box'' AI solutions. In particular, AI methods are known to amplify the bias in the data they are trained on, problematic in infectious disease models which leverage large open and inherently biased spatiotemporal data. These biases may propagate through the modeling pipeline to decision making, resulting in inequitable policy interventions. In this CSIRO-NSF collaborative project, our goal is to investigate how the AI disease modeling pipeline can lead to biased predictions and to derive solutions that mitigate this bias. Towards this goal, our team will 1) investigate an AI solution that captures spatio-temporal heterogeneity for infectious disease spread prediction using large sets of available human mobility and disease case data, 2) use agent-based simulation as a sandbox to control the bias in a variety of data sets, 3) evaluate how existing solution for fairness in AI are able to mitigate this bias and investigate novel solutions for fairness in AI. The team will apply the investigated AI solution to predict spread of emerging infectious diseases to support epidemiologists and decision-makers.