Automated Modeling of Clinical Narrative with High Definition Natural Language Processing Using Solor and Analysis Normal Form

• DOI: 10.3233/shti210822
• Publication Date: 2021-11-19T08:58:30Z
• Authors: Melissa P. Resnick; Frank LeHouillier; Steven H. Brown; Keith E. Campbell; Diane Montella; Peter L. Elkin
• Keywords: Reusability; Statement (logic)

One important concept in informatics is data which meets the principles of Findability, Accessibility, Interoperability and Reusability (FAIR). Standards, such as terminologies (findability), assist with tasks like interoperability, Natural Language Processing (NLP) (accessibility) decision support (reusability). terminology, Solor, integrates SNOMED CT, LOINC RxNorm. We describe HL7 Analysis Normal Form (ANF), their use high definition natural language processing (HD-NLP) program.We used HD-NLP to process 694 clinical narratives prior modeled by human experts into Solor ANF. compared output expert gold standard for 20% sample. Each statement was judged "correct" if matched ANF structure concepts, or "incorrect" any concepts were missing incorrect. Judgements summed give totals "incorrect".113 (80.7%) correct, 26 (18.6%) incorrect, 1 error. Inter-rater reliability 97.5% Cohen's kappa 0.948.The software provides useable complex standards-based representations statements designed drive CDS.