Feasibility of Privacy-Preserving Entity Resolution on Confidential Healthcare Datasets Using Homomorphic Encryption

TL;DR

This paper presents a privacy-preserving entity resolution method for healthcare datasets using homomorphic encryption, ensuring data privacy while maintaining accuracy and efficiency across large, confidential patient databases.

Contribution

The paper introduces AMPPERE, a novel cryptographic framework tailored for healthcare data, combining homomorphic encryption with parallelization for practical privacy-preserving entity resolution.

Findings

Effective in maintaining high accuracy in entity matching

Significantly improves computational efficiency over existing methods

Demonstrates scalability to large healthcare datasets

Abstract

Patient datasets contain confidential information which is protected by laws and regulations such as HIPAA and GDPR. Ensuring comprehensive patient information necessitates privacy-preserving entity resolution (PPER), which identifies identical patient entities across multiple databases from different healthcare organizations while maintaining data privacy. Existing methods often lack cryptographic security or are computationally impractical for real-world datasets. We introduce a PPER pipeline based on AMPPERE, a secure abstract computation model utilizing cryptographic tools like homomorphic encryption. Our tailored approach incorporates extensive parallelization techniques and optimal parameters specifically for patient datasets. Experimental results demonstrate the proposed method's effectiveness in terms of accuracy and efficiency compared to various baselines.