Infectious Disease Decision Support

This research targets one of the main barriers to the efficient monitoring and response to outbreaks, namely suboptimal and delayed decision-making, by providing new modes of decision support and integration of complex surveillance signals into action plans. Innovative analytic approaches using Bayesian classifiers and direct data based pattern recognition and clustering methods are applied to build rule-based decision support systems for clinical and public health assessments. This research also extends our current development of machine learning algorithms to provide patient-specific recommendations based on the molecular typing of bacteria with epidemic potential.