Community-Driven Terminology Mediation for Universal Health Exchange
|Customer||U.S. Army Medical Research and Materiel Command|
|Need||The most critical obstacle to seamless Health Information Exchange (HIE) is the translation from a proprietary medical terminology to universal medical standards. Stottler Henke is developing MediBridge, a new system for the US Army that will improve patient healthcare and support future health research.|
|Solution||Using a novel AI approach to terminology mapping, Medibridge will overcome the most critical obstacle to seamless Health Information Exchange (HIE): the translation from a proprietary medical terminology to universal medical standards. MediBridge lowers the barrier to HIE by exploiting feedback gathered from a community of domain and subdomain experts and pairing these community contributions with a committee of complementary “expert” evidence sources. An active learning module allows the system to recognize when ambiguity resolution by a human-in-the-loop may have a cascading effect throughout the automated mapping process. MediBridge then prioritizes those user operations that provide the biggest gain during the mapping process. By incorporating custom task-centered user interfaces into a semi-automatic terminology-mapping workflow, MediBridge enables creation of more comprehensive and precise mappings without imposing upon individual users.|
|Status||This project began in September 2014, and will end in February 2017.|
|Related Applications||The promise of MediBridge extends beyond just the medical domain. MediBridge demonstrates a new level of capability in the broader problem of ontology mapping—from innovations in semi-automated workflows to improved accuracy and scalability in identifying alignment points. By building custom tools tailored to users who are not experienced data modelers, the MediBridge semi-automatic workflow leads to more comprehensive and precise mappings while sparing users the usual burden of manual terminology mapping.|
This work is supported by the US Army Medical Research and Materiel Command under Contract No. W81XWH-13-C-0036.
The views, opinions and/or findings contained in this report are those of the author(s) and should not be construed as an official Department of the Army position, policy or decision unless so designated by other documentation.