nick pizzi home research publications miscellany

a public health decision support system framework

Advanced modelling technologies can support rapid and robust decision-making by public health experts developing effective strategies to mitigate the effects of infectious disease outbreaks. For instance, the agent-based model paradigm is a lattice-distributed collection of autonomous decision-making agents, the interactions of which unveil the dynamics and emergent properties of an infectious disease. Agent-based models permit effective representations of the complementary interactions between populations at a global level and individuals characterized by localized properties.

We are creating a comprehensive framework to implement these modelling technologies in order to design, develop and rapidly deploy decision support systems for the evaluation of infectious disease mitigation strategies. These systems, which must satisfy the criteria of reliability, computational efficiency, and adaptability to practical considerations and changes to fundamental knowledge, will make assumptions explicit, improve understanding of disease mechanisms, and provide sound predictions. This research is supported by NSERC.

pan-inform

Pan-InfORM (Pandemic Influenza Outbreak Research Modeling Team) is a multidisciplinary team evaluating mitigation strategies for pandemic preparedness in Canada, by employing computer models that integrate population demographics and mechanisms of infection control. With expertise in epidemiology, immunology, public health, software engineering, and mathematical modeling, we are currently investigating strategic response planning for emerging infectious diseases especially as they impact vulnerable populations.

predictive analytics

Computational systems that rely upon federated and disparate data repositories present significant data analysis challenges related to underlying reasoning, aggregation, and modelling processes. Users anticipate that “big data”, which are complex, dynamic, voluminous, and often unstructured, can be effectively accessed and summarized, with “interesting” findings revealed and communicated in forms that are easily understood. Compounding these issues are requirements such as privacy, temporal reasoning, information synchronization and interoperability, complex search queries, domain-relevant summaries, and secure data mining.

In close collaboration with our industrial partner, we are examining a comprehensive predictive analytics strategy to deal with the challenges and issues described above. We are investigating novel knowledge-based exploratory aggregation techniques operating on data with varying levels of information specificity as well as stochastic feature selection and data encoding methods that, in concert, identify data subsets with high information content while reducing input space dimensionality.

copyright © 2000–2017. n pizzi, phd