← All publications

Conference Paper

Enhancing Clinical Trial Data Queries with LLMs and Neo4j: A Flexible Framework for ADaM Dataset Management

Jaime Yan, Changhong Shi

PhUSE US Connect 2025 · 2025 · DH03

DOI · 10.5281/zenodo.22182899 Download PDF

Abstract

This paper models CDISC ADaM datasets in a Neo4j graph database via a reusable graph schema derived from ADaM specification files, and uses LLMs with LangChain to translate natural-language questions into both Cypher and SQL queries, making data checking more flexible, particularly for ad hoc tasks. Benchmarked head-to-head on 90 protocol-derived questions over synthetic ADaM datasets (ADSL, ADAE, ADCM, ADLB, ADVS) held in both PostgreSQL and Neo4j, LLM-generated Cypher achieved higher accuracy than SQL (73.33% vs 66.67% overall), with the advantage holding across difficulty tiers.

Keywords

Neo4j · Cypher · graph database · ADaM · LLM · SQL benchmarking · LangChain

Cite

Jaime Yan, Changhong Shi. "Enhancing Clinical Trial Data Queries with LLMs and Neo4j: A Flexible Framework for ADaM Dataset Management." PhUSE US Connect 2025, 2025. doi:10.5281/zenodo.22182899.