Abstract:Background Data sharing in multicenter medical research can improve the generalizability of research, accelerate progress, enhance collaborations among institutions, and lead to new discoveries from data pooled from multiple sources. Despite these benefits, many medical institutions are unwilling to share their data, as sharing may cause sensitive information to be leaked to researchers, other institutions, and unauthorized users. Great progress has been made in the development of secure machine learning frameworks based on homomorphic encryption in recent years; however, nearly all such frameworks use a single secret key and lack a description of how to securely evaluate the trained model, which makes them impractical for multicenter medical applications. Objective The aim of this study is to provide a privacy-preserving machine learning protocol for multiple data providers and researchers (eg, logistic regression). This protocol allows researchers to train models and then evaluate them on medical data from multiple sources while providing privacy protection for both the sensitive data and the learned model. Methods We adapted a novel threshold homomorphic encryption scheme to guarantee privacy requirements. We devised new relinearization key generation techniques for greater scalability and multiplicative depth and new model training strategies for simultaneously training multiple models through x-fold cross-validation. Results Using a client-server architecture, we evaluated the performance of our protocol. The experimental results demonstrated that, with 10-fold cross-validation, our privacy-preserving logistic regression model training and evaluation over 10 attributes in a data set of 49,152 samples took approximately 7 minutes and 20 minutes, respectively. Conclusions We present the first privacy-preserving multiparty logistic regression model training and evaluation protocol based on threshold homomorphic encryption. Our protocol is practical for real-world use and may promote multicenter medical research to some extent.

Secure Multi-pArty Computation Grid LOgistic REgression (SMAC-GLORE)

A Hybrid-Domain Framework for Secure Gradient Tree Boosting.

Secret Sharing based Secure Regressions with Applications

High Performance Logistic Regression for Privacy-Preserving Genome Analysis

Fair compute loads enabled by blockchain: sharing models by alternating client and server roles

Privacy-preserving two-parties logistic regression on vertically partitioned data using asynchronous gradient sharing

Hybrid RAG-empowered Multi-modal LLM for Secure Data Management in Internet of Medical Things: A Diffusion-based Contract Approach

Supporting Regularized Logistic Regression Privately and Efficiently

Web-Based Privacy-Preserving Multicenter Medical Data Analysis Tools Via Threshold Homomorphic Encryption: Design and Development Study

Privacy-preserving multi-party logistic regression in cloud computing

Multicenter Privacy-Preserving Cox Analysis Based on Homomorphic Encryption

Efficient Privacy Preserving Logistic Regression for Horizontally Distributed Data

Online Efficient Secure Logistic Regression based on Function Secret Sharing

SecureML: A System for Scalable Privacy-Preserving Machine Learning

Learning from electronic health records across multiple sites: A communication-efficient and privacy-preserving distributed algorithm

Privacy-preserving logistic regression with secret sharing

Enabling End-to-End Secure Federated Learning in Biomedical Research on Heterogeneous Computing Environments with APPFLx

Secure multi-party linear regression at plaintext speed

Large-scale Secure XGB for Vertical Federated Learning

Secure Collaborative Training and Inference for XGBoost

Privacy Preserving Multi-Agent Reinforcement Learning in Supply Chains