Privacy-Preserving Distributed Deep Learning for Clinical Data

TL;DR

This paper introduces a differential privacy-based method for distributed deep learning on clinical data, enabling multiple institutions to collaboratively train models without compromising patient privacy.

Contribution

It proposes a novel approach combining distributed training with differential privacy guarantees specifically for sensitive clinical datasets.

Findings

Effective privacy preservation demonstrated on real clinical datasets

Models trained with privacy guarantees maintain high accuracy

Applicable to multi-institutional collaborations in healthcare

Abstract

Deep learning with medical data often requires larger samples sizes than are available at single providers. While data sharing among institutions is desirable to train more accurate and sophisticated models, it can lead to severe privacy concerns due the sensitive nature of the data. This problem has motivated a number of studies on distributed training of neural networks that do not require direct sharing of the training data. However, simple distributed training does not offer provable privacy guarantees to satisfy technical safe standards and may reveal information about the underlying patients. We present a method to train neural networks for clinical data in a distributed fashion under differential privacy. We demonstrate these methods on two datasets that include information from multiple independent sites, the eICU collaborative Research Database and The Cancer Genome Atlas.