Bridging the Gap to Universal Health Information Exchange with Community-Driven Terminology Mediation

The goal of this proposed effort is to develop a new system for the US Army to improve patient healthcare and support future health research by lowering barriers to Health Information Exchange (HIE). The most critical obstacle to seamless HIE is the translation from a proprietary medical terminology to a universal reference language (e.g., universal standards). Our system, called MediBridge, addresses this challenge through three primary techniques. First, we employ a novel approach to terminology mapping by exploiting feedback gathered from a community of domain and subdomain experts. Second, we will enhance mapping accuracy by pairing these community contributions with a committee of complementary “expert” evidence sources. Finally, we will incorporate an active learning module to recognize when ambiguity resolution by a human-in-the-loop can have a cascading effect throughout the automated mapping process and thereby prioritize user operations that can profitably constrain/direct the process with little effort. We established the feasibility of our approach by prototyping custom task-centered user interfaces within a semiautomatic terminology-mapping workflow in order to create more comprehensive and precise mappings without imposing upon individual users. Phase I proved the concept and has set the stage for full implementation of MediBridge in Phase II.