# RedactOR: An LLM-Powered Framework for Automatic Clinical Data De-Identification

arXiv: https://arxiv.org/abs/2505.18380 · HTML: https://mlsanity.com/paper/2505.18380/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 8 of its 42 references that are in the map.
- First posted on arXiv 23 May 2025.

## Builds on (most cited first, 8 of 8)

- [End-to-end Sequence Labeling via Bi-directional LSTM-CNNs-CRF](https://mlsanity.com/paper/1603.01354/) (2016)
- [DeID-GPT: Zero-shot Medical Text De-Identification by GPT-4](https://mlsanity.com/paper/2303.11032/) (2023)
- [LeanContext: Cost-Efficient Domain-Specific Question Answering Using LLMs](https://mlsanity.com/paper/2309.00841/) (2023)
- [Towards Optimizing the Costs of LLM Usage](https://mlsanity.com/paper/2402.01742/) (2024)
- [NeuroNER: an easy-to-use program for named-entity recognition based on neural networks](https://mlsanity.com/paper/1705.05487/) (2017)
- [De-identification of clinical free text using natural language processing: A systematic review of current approaches](https://mlsanity.com/paper/2312.03736/) (2023)
- [A Comparative Evaluation Of Transformer Models For De-Identification Of Clinical Text Data](https://mlsanity.com/paper/2204.07056/) (2022)
- [Can Zero-Shot Commercial APIs Deliver Regulatory-Grade Clinical Text DeIdentification?](https://mlsanity.com/paper/2503.20794/) (2025)

## Built on by (most cited first, 5 of 5)

- [Towards Context-Aware Image Anonymization with Multi-Agent Reasoning](https://mlsanity.com/paper/2603.27817/) (2026)
- [SHIELD: A Diverse Clinical Note Dataset and Distilled Small Language Models for Enterprise-Scale De-identification](https://mlsanity.com/paper/2605.03301/) (2026)
- [Institution-Specific LLM Prompting Recovers PHI That De-identification Systems and Their Gold Standards Both Miss](https://mlsanity.com/paper/2608.17051/) (2026)
- [Consistent Relexicalization of Clinical Documents using Graph-Based Approach](https://mlsanity.com/paper/2609.21387/) (2026)
- [AegisVox: Factorized Privacy and Clinical Preservation for Long-Form Speech](https://mlsanity.com/paper/2610.14643/) (2026)

## Abstract (arXiv)

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.