Natural Language Processing to Identify Unmet Needs in Military Medicine

The U.S. Army Telemedicine and Advanced Technology Research Center (TATRC) Advanced Medical Technology Initiative (AMTI) seeks to identify and demonstrate key emerging technologies related to military medicine. AMTI invites researchers to submit proposals for short-term funding opportunities that support this goal. The Extended Innovation Fund (EIF) is one such opportunity that provides funding for an 18-month period of performance. EIF proposals contain several prompted free-text descriptions of the research to be performed. These free-text descriptions contain insights regarding research in military medicine that could inform military leadership of unmet research needs, however manual review of all historic proposals for common research themes is labor intensive. To address this issue, we developed and applied a process to extract and summarize key proposal themes leveraging natural language processing and unsupervised machine learning. The result of this process were proposal categories for the problem proposed to be address (“problem-sets”) and for the proposed technological solution (“solution-sets”). The focus of this presentation are the proposal solution-sets related to health information technology.