RedactOR: An LLM-Powered Framework for Automatic Clinical Data De-Identification
cs.AI, cs.LG
Submitted: 2025-05-23
Updated: 2025-07-24
Comments: Accepted to ACL 2025 Industry Track. To appear
Journal ref: Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025), Volume 6: Industry Track, pages 510-530, 2025
DOI: 10.18653/v1/2025.acl-industry.36
License: http://creativecommons.org/licenses/by-sa/4.0/
The gist: Ensuring clinical data privacy while preserving utility is critical for AI-driven healthcare and data analytics.
Terminology
Abstract
Ensuring clinical data privacy while preserving utility is critical for AI-driven healthcare and data analytics. Existing de-identification (De-ID) methods, including rule-based techniques, deep learning models, and large language models (LLMs), often suffer from recall errors, limited generalization, and inefficiencies, limiting their real-world applicability. We propose a fully automated, multi-modal framework, RedactOR for de-identifying structured and unstructured electronic health records, including clinical audio records. Our framework employs cost-efficient De-ID strategies, including intelligent routing, hybrid rule and LLM based approaches, and a two-step audio redaction approach. We present a retrieval-based entity relexicalization approach to ensure consistent substitutions of protected entities, thereby enhancing data coherence for downstream applications. We discuss key design desiderata, de-identification and relexicalization methodology, and modular architecture of RedactOR and its integration with the Oracle Health Clinical AI system. Evaluated on the i2b2 2014 De-ID dataset using standard metrics with strict recall, our approach achieves competitive performance while optimizing token usage to reduce LLM costs. Finally, we discuss key lessons and insights from deployment in real-world AI- driven healthcare data pipelines.
Sources
- NeuroNER: an easy-to-use program for named-entity recognition based on neural networks
- DeID-GPT: Zero-shot Medical Text De-Identification by GPT-4
- End-to-end Sequence Labeling via Bi-directional LSTM-CNNs-CRF
- A Comparative Evaluation Of Transformer Models For De-Identification Of Clinical Text Data
- Doing More with Less: A Survey on Routing Strategies for Resource Optimisation in Large Language Model-Based Systems
- Towards Optimizing the Costs of LLM Usage
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems
- VSAL: A Vision Solver with Adaptive Layouts for Graph Property Detection