Differential Privacy-enabled Federated Learning for Sensitive Health Data

Abstract

Leveraging real-world health data for machine learning tasks requires addressing many practical challenges, such as distributed data silos, privacy concerns with creating a centralized database from person-specific sensitive data, resource constraints for transferring and integrating data from multiple sites, and risk of a single point of failure. In this paper, we introduce a federated learning framework that can learn a global model from distributed health data held locally at different sites. The framework offers two levels of privacy protection. First, it does not move or share raw data across sites or with a centralized server during the model training process. Second, it uses a differential privacy mechanism to further protect the model from potential privacy attacks. We perform a comprehensive evaluation of our approach on two healthcare applications, using real-world electronic health data of 1 million patients. We demonstrate the feasibility and effectiveness of the federated learning framework in offering an elevated level of privacy and maintaining utility of the global model.

Machine Learning for Health (ML4H) at NeurIPS 2019

References in corpus (2)

• Differential Privacy and Machine Learning: a Survey and Review
• Anonymizing Data for Privacy-Preserving Federated Learning

Cited by in corpus (11)

• Local Differential Privacy and Its Applications: A Comprehensive Survey
• Federated learning for violence incident prediction in a simulated cross-institutional psychiatric setting
• Wireless Federated Learning with Local Differential Privacy
• Understanding Clipping for Federated Learning: Convergence and Client-Level Differential Privacy
• A Systematic Literature Review on Federated Learning: From A Model Quality Perspective
• Personalized Federated Deep Learning for Pain Estimation From Face Images
• Evaluating Progress on Machine Learning for Longitudinal Electronic Healthcare Data
• Exploring privacy-enhancing technologies in the automotive value chain
• Anonymizing Data for Privacy-Preserving Federated Learning
• Failure Prediction in Production Line Based on Federated Learning: An Empirical Study
• Optimizing the Numbers of Queries and Replies in Federated Learning with Differential Privacy