Semantic Modeling of Cohort Descriptions in Research Studies

Recommendations in ADA’s Standards of Medical Care in Diabetes guideline are supported by findings from scientific publications (primarily clinical trials and case studies). We propose an approach rooted in Information Extraction and Knowledge Representation techniques to augment guideline representations with population descriptions from cited literature. We develop a Cohort Ontology, modeling patient groups and interventions mentioned in cited publications, ultimately aiming at assisting physicians and computers to suggest recommendations for complex patients using cohort alignment.

View Publication

Associated Projects

The Center for Health Empowerment by Analytics, Learning, and Semantics (HEALS) is a five-year collaboration between Rensselaer and IBM aimed at researching how the application of advanced cognitive computing capabilities can help people to understand and improve their own health conditions.