SBIR 136 - OMNISYNC: A Real-Time Comprehensive Knowledge Graph for the Biomedical Science Community

NIH RePORTER · NIH · N43 · $299,957 · view on reporter.nih.gov ↗

Abstract

The modern biomedical landscape is punctuated by a surge of innovative findings, a substantial chunk of which resides in academic papers and esteemed journal publications. While these documents harbor groundbreaking results, their intricate and textual nature often leaves the knowledge latent and underutilized. Our proposed methodology aims to bridge this gap by representing these critical insights as structured Knowledge Graphs (KGs). The end-product proposed—a dynamically enriched biomedical Knowledge Graph (KG) integrated with the latest research insights and powered by advanced machine learning models—offers several significant advantages over current methodologies and technologies: Successful completion of this technical objective would prove that training data can be vastly expanded based on a huge corpus of academic works that we have already indexed in a vector database.

Key facts

NIH application ID
11214915
Project number
75N93024C00037-0-9999-1
Recipient
OMNISYNC INCORPORATED
Principal Investigator
NORMAN HUANG
Activity code
N43
Funding institute
NIH
Fiscal year
2024
Award amount
$299,957
Award type
Project period
2024-09-05 → 2025-09-04