Seminar from the Health eResearch Centre
|Starts:||12:00 23 May 2018|
|Ends:||13:00 23 May 2018|
|What is it:||Seminar|
|Organiser:||Faculty of Biology, Medicine and Health|
|Who is it for:||University staff, Adults, Current University students|
Prof John Holmes, Professor of Medical Informatics in Epidemiology at the University of Pennsylvania Perelman School of Medicine
Abstract: Purely clinical data, as is captured in the electronic medical record, is demonstrably incomplete when one considers the impact of environmental exposures on health. Such exposures are very seldom captured by practitioners, and even then, usually incompletely. Increasingly, researchers and data scientists are becoming interested in methods for integrating many different types of data from many different sources into a single, analytic data resource that can be used to evaluate the associations of environmental, demographic, social, historical, and clinical features with an individual’s health current and possibly future health status.
However, environmental data are extremely complex and at best, provide ecological associations. These data would include assessments of an individual’s built environment (housing, neighbourhood characteristics), social environment (connectedness, social capital), and weather, to name just three. However, such data are available via various agencies that keep records of these phenomena. Bringing these data together with the clinical is difficult, but can be facilitated through the create of computable ontologies that resolve differences in variable representations, measurement scales, and exposure windows. We will describe our experience in using ontology-driven integration of heterogeneous data to create useful analytic datasets that include comprehensive environmental features for exposomic research.
Role: Professor of Medical Informatics in Epidemiology
Organisation: University of Pennsylvania Perelman School of Medicine
Travel and Contact Information
Lecture Theatre 4