Tribal Abstraction Networks for Hierarchical Ontology and Terminology Structures


Tribal Abstraction Networks for Hierarchical Ontology and Terminology Structures

NJIT Case No. 14-010


Inventors: James Geller, Yehoshua Perl, Christopher Ochs


Intellectual Property & Development status: Patent Publication No. US20170039295A1. US Patent Protection is pending.

NJIT is currently seeking commercial partners for the further development and commercialization of this opportunity.


Technology Brief: Researchers at New Jersey Institute of Technology in the Department of Computer Science have invented a novel solution for deriving and visualizing an abstraction network for SNOMED CT ( The Systematized Nomenclature of Medicine – Clinical Terms) that do not have semantic relationship as well.


SNOMED CT provides the core general terminology for electronic health records. The terminologies are large and complex due to which errors are hidden. Such errors can be identified by deriving abstraction network from a large terminology. An abstraction network is a high-level, simplified, view of a complex structure. Visualizations for such abstraction networks aids auditors to quickly focus on elements of interest within a terminology. Previous work regarding abstraction networks assumed that terminologies (SNOMED CT) have semantic (i.e., lateral, IS-A e.g: “diabetes mellitus type 2”  IS-A “diabetes mellitus”) relationships. However, there are many terminologies, including 12 hierarchies within SNOMED CT that do not have semantic relationships. This invention addresses this problem by deriving and visualizing an abstraction network even when no semantic relationships are present. The invention defines groups of classes (i.e. concepts) called tribes by using the descendants of each child of the root in a hierarchy. The tribal abstraction network captures the intersection sets of the tribes, designing a tribal network.

Published Patent is available at


Inventors Bio:

James Geller, is a Professor and former Chair of the Computer Science Department New Jersey Institute of Technology. He currently serves as Associate Dean for Research of the College of Computing Sciences at NJIT. He received his PhD from the State University of New York at Buffalo in 1988 in Computer Science, with a focus on AI/Knowledge Representation. Dr. Geller cofounded SABOC (the Structural Analysis of Biomedical Ontologies Center) at CS/NJIT and founded the Semantic Web Laboratory at the Computer Science Department of NJIT. He has published over 190 journal and conference papers and 14 book chapters in Medical Informatics, Semantic Web Technology, Object-Oriented Database Modeling, Knowledge Representation, etc. Between 2006 and 2012 Dr. Geller was Co-Principal Investigator on several federal grants, totaling over $2,500,000, on auditing methods, abstraction algorithms and software tools for important medical terminology systems such as the Unified Medical Language System (UMLS) and the Systematized Nomenclature of Medicine – Clinical Terms (SNOMED CT). From 2012-2015 he was investigator on an NSF grant for teaching Cyber Security (iSECURE). Currently he is co-PI on a major NIH grant on family-based quality assurance for Biomedical Ontologies. In 2012 Dr. Geller was inducted as a Fellow of the American College of Medical Informatics (ACMI).

Dr. Geller received the NJIT Master Teacher Designation (2005) and three other NJIT teaching awards in 2002, 2003 and 2011. Dr. Geller also received an NJIT College Research Award (2010).



Patent Information:
For Information, Contact:
Simon Nynens
VP, Business Incubation
New Jersey Institute of Technology
James Geller
Yehoshua Perl
Christopher Ochs
Patent Pending
© 2019. All Rights Reserved. Powered by Inteum