MainSEL

MainSEL performs privacy-preserving record linkage (PPRL) across medical and other sensitive datasets using secure multi-party computation (SMC) to link records without revealing personally identifiable information (PII).


Key Features:

  • Privacy-Preserving Record Linkage: Enables linkage of medical databases and other sensitive datasets at the record level even in the absence of exact identifiers such as names or dates of birth.
  • Secure Multi-Party Computation (SMC): Employs SMC techniques so parties can compute matches without exposing PII and without relying on a trusted third party.
  • Fault-Tolerant Framework: Provides robust performance under challenging network conditions with benchmarks reporting ~48 seconds to link a patient record against 10,000 records over a 100 ms delayed connection and ~3.9 seconds with low-latency connections.
  • Integration with Mainzelliste: Built on the medical record-keeping software Mainzelliste to support secure linkage tasks while preserving data integrity and privacy.

Scientific Applications:

  • Healthcare and Biomedical Research: Facilitates integration of patient records from multiple institutions to support comprehensive data analysis while protecting PII.
  • Multi-Institutional Collaboration: Enables linkage across institutions to conduct large-scale studies and combine disparate datasets without exposing sensitive identifiers.

Methodology:

MainSEL applies cryptographic techniques and secure computation protocols, implementing secure multi-party computation to perform computations on encrypted data without revealing sensitive information and eliminating the need for a trusted third party.

Topics

Details

License:
AGPL-3.0
Tool Type:
command-line tool
Programming Languages:
Java
Added:
1/18/2021
Last Updated:
11/24/2024

Operations

Publications

Stammler S, Kussel T, Schoppmann P, Stampe F, Tremper G, Katzenbeisser S, Hamacher K, Lablans M. Mainzelliste SecureEpiLinker (MainSEL): privacy-preserving record linkage using secure multi-party computation. Bioinformatics. 2020;38(6):1657-1668. doi:10.1093/bioinformatics/btaa764. PMID:32871006. PMCID:PMC8896632.

PMID: 32871006
PMCID: PMC8896632
Funding: - German Federal Ministry of Education and Research (BMBF) through the HiGHmed Consortium: 01ZZ1802G - German Research Foundation (DFG) through the MAGIC project: LA 3859/1-1 - Research Training Group GRK: 1651