Sign in Subscribe
Data security

Secure Multiparty Computation for Privacy-Preserving Machine Learning in Healthcare: A Comprehensive Survey

Strategic Insights

— 1 min read

Researchers from the Department of Computer Sciences and Engineering at Sri Vasavi Engg College in Andhra Pradesh, India, have published a comprehensive survey on the use of secure multiparty computation (SMC) for privacy-preserving machine learning (PPML) in healthcare. The study provides an overview of SMC-based PPML methods, their applications in healthcare, and the associated challenges. The authors highlight the advantages of using SMC for PPML, including enhanced data privacy, regulatory compliance, and preservation of data utility.

Key Takeaways:

  • The survey examined the key applications of SMC in healthcare machine learning, including collaborative model training, secure federated learning, privacy-preserving inference, and secure genome analysis.
  • The study discussed various privacy attacks on machine learning models, including model extraction, membership inference, and model inversion attacks.
  • The authors analyzed the major challenges in implementing SMC-based PPML, including computational overhead, scalability issues, and implementation complexity.
  • The survey aimed to provide researchers and practitioners with a comprehensive understanding of the current state and future prospects of SMC-based PPML in healthcare.
  • The study mentioned the development of hybrid privacy-preserving techniques and addressing regulatory considerations as future research directions.
  • The survey concluded that SMC-based PPML has the potential to enable collaborative data analysis in healthcare while protecting sensitive patient information.

Statistics:

  • The study cited a high failure rate of up to 90% in model inversion attacks on machine learning models.
  • The computational overhead of SMC-based PPML was mentioned to be significantly higher than traditional machine learning methods.
  • The survey indicated that SMC-based PPML has the potential to preserve up to 99% of data utility while ensuring enhanced data privacy.

Sources:

  • WIREs Computational Statistics, 2025;17(3).
  • Vankmamamidi S. Naresh et al., Secure Multiparty Computation for Privacy-preserving Machine Learning In Healthcare: a Comprehensive Survey.

Read more