Logo image
Back
Governance-Aware, Privacy-Preserving Retrieval-Augmented Large Language Models for Clinical Triage Decision Support
Conference proceeding   Peer reviewed

Governance-Aware, Privacy-Preserving Retrieval-Augmented Large Language Models for Clinical Triage Decision Support

ABM Kamrul Islam Riad, Md Abdul Barek, Hossain Shahriar, Guillermo Francia III, Sheikh Iqbal Ahamed, Eman El-Sheikh and Yasmeen Rawajfih
2026 IEEE International Conference on Digital Health (ICDH), pp.23-32
International Conference on Digital Health (ICDH) (Sydney, Australia, 07/13/2026–07/18/2026)
07/13/2026
DOI: https://doi.org/10.1109/ICDH72779.2026.00012

Metrics

1 Record Views

Abstract

Agentic AI Clinical Decision Support Differential Privacy Explainable AI (XAI) Large language models PrivacyUtility Trade-off Retrieval-Augmented Generation (RAG) hallucinations Artificial Intelligence or Cybernetics Electronic Health Records Intelligent Agents or Systems
Large Language Models (LLMs) offer significant potential for clinical decision support, but their deployment is constrained by hallucination, privacy risks, and limited interpretability. Retrieval-Augmented Generation (RAG) improves reliability by grounding outputs in patient-specific electronic health records (EHRs); however, it introduces new challenges related to sensitive data exposure. This paper proposes a governanceaware, privacy-preserving RAG framework for clinical triage decision support using the MIMIC-IV-Ext dataset. The framework integrates retrieval-based grounding, differential privacy, explainability, and an agentic orchestration layer within a unified architecture. An ∊ -ledger is introduced to track cumulative privacy expenditure, and a Privacy-Utility-Explainability (PUE) framework is used to systematically evaluate trade-offs. Experimental results demonstrate that the proposed approach improves classification performance (e.g., achieving up to 82% accuracy and 88% AUROC) and reduces under-triage risk, while privacy mechanisms introduce only modest performance degradation. Retrieval-based attribution further enhances interpretability by linking predictions to clinically relevant cases. These results highlight the effectiveness of combining retrieval, privacy, and explainability for developing trustworthy and deployable healthcare AI systems.

Details

Logo image