Reliable Curation of EHR Dataset via Large Language Models under Environmental Constraints

TL;DR

CELEC tackles the barrier that researchers face in querying electronic health records by providing a privacy-preserving, LLM-powered NL-to-SQL system. It relies on schema-grounded prompting with few-shot demonstrations and chain-of-thought reasoning, and executes SQL locally while exposing only metadata to the LLM. On the adapted EHRSQL benchmark, it achieves competitive $RS(0)$ performance with a score of $81.05\%$ and a per-query latency of about $6$ seconds, without any training data. The approach also includes an LLM-driven visualization module to aid exploratory analysis, illustrating a practical route to accelerate biomedical discovery while maintaining stringent privacy. Overall, CELEC demonstrates that strong data analytics on EHRs can be made accessible to researchers without compromising data security or requiring specialized database expertise.

Abstract

Electronic health records (EHRs) are central to modern healthcare delivery and research; yet, many researchers lack the database expertise necessary to write complex SQL queries or generate effective visualizations, limiting efficient data use and scientific discovery. To address this barrier, we introduce CELEC, a large language model (LLM)-powered framework for automated EHR data extraction and analytics. CELEC translates natural language queries into SQL using a prompting strategy that integrates schema information, few-shot demonstrations, and chain-of-thought reasoning, which together improve accuracy and robustness. On a subset of the EHRSQL benchmark, CELEC achieves execution accuracy comparable to prior systems while maintaining low latency, cost efficiency, and strict privacy by exposing only database metadata to the LLM. CELEC also adheres to strict privacy protocols: the LLM accesses only database metadata (e.g., table and column names), while all query execution occurs securely within the institutional environment, ensuring that no patient-level data is ever transmitted to or shared with the LLM. Ablation studies confirm that each component of the SQL generation pipeline, particularly the few-shot demonstrations, plays a critical role in performance. By lowering technical barriers and enabling medical researchers to query EHR databases directly, CELEC streamlines research workflows and accelerates biomedical discovery.

Figures (2)

• Figure 1: Overview of the CELEC system design
• Figure 2: CELEC user interface, with results of running on an example question