LLM-assisted approach reconciles EHR and clinical note medication data

Researchers have developed a novel approach to reconcile discrepancies between medication information found in clinical notes and structured electronic health records (EHRs). Their method employs large language models (LLMs) for reference construction, combined with human review and deterministic normalization, to compare semantic and temporal data from both sources. The study found that while exact matches between note-derived and structured medication history were infrequent, semantic overlap significantly increased when considering broader definitions and timeframes, highlighting the importance of advanced normalization techniques for accurate patient medication profiles. AI

IMPACT Improves accuracy of patient medication data by reconciling disparate EHR sources using LLMs.

RANK_REASON The cluster contains an academic paper detailing a new methodology for data reconciliation. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CL →

• electronic health records
• Observational Medical Outcomes Partnership Common Data Model

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →