Privacy-Preserving Integration of Medical Data : A Practical Multiparty Private Set Intersection.

Area of Science:

• Computer Science
• Data Security
• Cryptography

Background:

• Medical data is often siloed across different organizations, hindering comprehensive analysis.
• Integrating these datasets is crucial for research but must preserve patient and commercial privacy.
• Multiparty Private Set Intersection (MPSI) is a privacy-preserving protocol for finding common data points across multiple datasets.

Purpose of the Study:

• To propose a practical Multiparty Private Set Intersection (MPSI) protocol.
• To ensure dataset size and computational complexity are independent of the number of participating parties.
• To reduce computational overhead by utilizing an outsourcing provider unaware of the data.

Main Methods:

• Developed a novel MPSI protocol leveraging an oblivious outsourcing provider.
• Implemented a prototype on a virtual private network to facilitate parallel computation.
• Evaluated the performance of the proposed MPSI protocol through empirical testing.

Main Results:

• The proposed MPSI protocol achieves independence in dataset size and per-party computational complexity.
• The use of an outsourcing provider significantly reduces computational load.
• The implemented prototype demonstrated superior efficiency compared to existing MPSI approaches.

Conclusions:

• The developed MPSI protocol offers a practical and efficient solution for secure data integration.
• This approach effectively balances the need for detailed data analysis with stringent privacy requirements.
• The protocol's efficiency and scalability make it suitable for real-world applications involving sensitive data.