Home // eTELEMED 2020, The Twelfth International Conference on eHealth, Telemedicine, and Social Medicine // View article

Privacy Preserving Fuzzy Patient Matching Using Homomorphic Encryption

Authors:
Shiva Ashish Thumparthy
Ilya Sher

Keywords: record linkage; homomorphic encryption; ciphertext packing; Bloom filter.

Abstract:
Patient record linkage is an important operation that is necessary for identifying similar patients across medical facilities, with the ambition to improve patient outcomes. However, with increased data privacy concerns, these record linkage algorithms must protect sensitive patient demographic information. Previous works have included usage of Bloom filters (susceptible to frequency attacks) and homomorphic encryption (high computational complexity and communication overheads). We propose a record linkage algorithm that utilizes fully homomorphic encryption ciphertext packing for matrices. This ensures that the algorithm remains privacy-preserving and resilient to multiple attack vectors, while allowing multiple patients’ records to be compared at once, as opposed to pairwise comparison.

Pages: 85 to 86

Copyright: Copyright (c) IARIA, 2020

Publication date: March 22, 2020

Published in: conference

ISSN: 2308-4359

ISBN: 978-1-61208-763-4

Location: Valencia, Spain

Dates: from November 21, 2020 to November 25, 2020