Privacy-preserving patient clustering for personalized federated learning

TL;DR

This paper introduces PCBFL, a privacy-preserving clustered federated learning framework for medical data that improves patient cohorting and model performance while safeguarding patient privacy using cryptographic techniques.

Contribution

The paper presents PCBFL, a novel cryptographic approach enabling patient-level clustering in federated learning without compromising privacy, specifically tailored for medical applications.

Findings

PCBFL forms meaningful patient risk cohorts.

It improves model performance with an average AUC increase of 4.3%.

It enhances AUPRC by 7.8% over existing methods.

Abstract

Federated Learning (FL) is a machine learning framework that enables multiple organizations to train a model without sharing their data with a central server. However, it experiences significant performance degradation if the data is non-identically independently distributed (non-IID). This is a problem in medical settings, where variations in the patient population contribute significantly to distribution differences across hospitals. Personalized FL addresses this issue by accounting for site-specific distribution differences. Clustered FL, a Personalized FL variant, was used to address this problem by clustering patients into groups across hospitals and training separate models on each group. However, privacy concerns remained as a challenge as the clustering process requires exchange of patient-level information. This was previously solved by forming clusters using aggregated data, which led to inaccurate groups and performance degradation. In this study, we propose Privacy-preserving Community-Based Federated machine Learning (PCBFL), a novel Clustered FL framework that can cluster patients using patient-level data while protecting privacy. PCBFL uses Secure Multiparty Computation, a cryptographic technique, to securely calculate patient-level similarity scores across hospitals. We then evaluate PCBFL by training a federated mortality prediction model using 20 sites from the eICU dataset. We compare the performance gain from PCBFL against traditional and existing Clustered FL frameworks. Our results show that PCBFL successfully forms clinically meaningful cohorts of low, medium, and high-risk patients. PCBFL outperforms traditional and existing Clustered FL frameworks with an average AUC improvement of 4.3% and AUPRC improvement of 7.8%.