Proto-OKN Theme 1: A Dynamically-Updated Open Knowledge Network for Health: Integrating Biomedical Insights with Social Determinants of Health

University Of Virginia Main Campus, Charlottesville VA

Investigators

Abstract

This project aims to expand current biomedical knowledge graphs by incorporating Social Determinants of Health (SDoH) data, which are typically underutilized despite their proven connection to health outcomes. The goal is to create a comprehensive health knowledge graph that combines biomedical facts and SDoH data from scientific literature and Electronic Health Records. Crucially, this model will allow for continuous updates from ongoing streams of biomedical and clinical text data, promoting swift distribution of new research findings and fostering collaboration in the ever-evolving biomedical and public health fields. Biomedical knowledge graphs (KGs), which organize, analyze, and apply biomedical information, rarely incorporate data related to non-clinical factors like socioeconomic status, education, and employment, that have proven correlations with several health outcomes. It is crucial to integrate SDoH into open knowledge repositories to enhance health improvement efforts and to reduce persistent disparities in healthcare resource accessibility. One outcome of this integration would be a comprehensive knowledge network for health that merges biomedical facts with SDoH data. This network could continually update itself with new data. Challenges to this endeavor include data integration, the dynamic nature of data streams, and the need to combine biomedical knowledge with SDoH data in an equitable and ethical manner. To address these challenges, the project will create a multi-dimensional knowledge network capable of supporting complex queries across various applications and automatically verifying information quality. The project will also link the KGs with electronic health records to uncover associations between social determinants and health outcomes, thereby improving healthcare outcomes and promoting health equity. This project innovates by advancing semantic interoperability, improving knowledge representation, enabling adaptive knowledge acquisition, ensuring knowledge graph trustworthiness, and promoting ethical awareness. It aims to yield a comprehensive knowledge network, open-source algorithms, models, and tools that integrate SDoH facts and data. The accessibility of the tools created in this project will accelerate knowledge discovery, improving the understanding of human health and the wellbeing of vulnerable populations. Additionally, this project will significantly contribute to the education of students, particularly those from underrepresented groups, through special seminar courses and ongoing diversity initiatives. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.