Large Language Models for Structured Clinical Data Analysis: Dual-Agent Grounding and Validation

2026-09-28

AI Summary

This study addresses the hallucination and untraceability issues of large language models in structured clinical data analysis by proposing CLEAR-Med, a dual-agent framework. The method employs a bounded selective pipeline architecture that decouples SQL generation from independent verification, introducing a cross-provider validation agent alongside deterministic logic checks while supporting a fail-closed mechanism to ensure safety. Evaluated on a neonatal hypoxic-ischemic encephalopathy (HIE) dataset, the proposed framework achieves an accuracy of 66.4%, representing a 54.4 percentage point improvement over baseline methods. These results demonstrate that CLEAR-Med enables traceable and highly reliable natural language analysis of clinical data, effectively mitigating the inherent limitations of standard large language model pipelines in safety-critical healthcare applications.

Abstract

Objective: To develop and characterize CLEAR-Med, a dual-agent framework for natural-language analysis of structured clinical data that separates SQL-based invocation from independent validation. Methods: CLEAR-Med uses one agent to translate a question into executable Structured Query Language (SQL), retain the executed query and database result, and produce a draft. Deterministic checks and a separately invoked cross-provider Validation Agent then accept the draft, request one bounded repair, or abstain. We formalized the system as a bounded selective pipeline and evaluated CLEAR-Med's configuration and scalability, and the Invocation Agent's accuracy and consistency on a 25-query development benchmark, using a harmonized 21-site neonatal hypoxic-ischemic encephalopathy table containing 532 de-identified infant records and approximately 1,300 variables. Results: CLEAR-Med completed all six nominal scalability configurations, including 500x1300. Across 25 development-benchmark queries repeated five times, the Invocation Agent answered 83 of 125 responses correctly (66.4%; query-cluster bootstrap 95% CI, 48.0-83.2%), compared with 15 of 125 (12.0%; 95% CI, 3.2-22.4%) for the ungrounded ChatGPT baseline, a paired improvement of 54.4 percentage points (95% CI, 36.8-72.0%). Conclusion: CLEAR-Med provides a general architecture for traceable analysis of structured clinical data: numerical claims remain linked to executed SQL, and unresolved cases can fail closed. The reported experiments characterize CLEAR-Med's configuration and scalability and the Invocation Agent's accuracy, while the formal analysis establishes the encoded-property guarantee of the complete control flow; a prospective full-pipeline evaluation of the validation and abstention stages is the next stage of this work.

Problem

• Large Language Models
• Structured Clinical Data
• Natural Language Analysis
• Data Grounding
• Validation

Innovation

• Dual-Agent Framework
• Structured Clinical Data
• SQL Grounding
• Bounded Selective Pipeline
• Validation Agent