FLOP: Federated Learning on Medical Datasets using Partial Networks

TL;DR

FLOP introduces a federated learning approach that shares only partial models to enhance privacy and security in medical data analysis, demonstrating effective COVID-19 diagnosis without data sharing.

Contribution

The paper proposes FLOP, a novel federated learning algorithm that shares partial networks, improving privacy and security in collaborative medical data modeling.

Findings

Achieves comparable or better performance than traditional FL methods.

Reduces privacy and security risks by sharing only partial models.

Enables hospitals to collaboratively train models without sharing sensitive patient data.

Abstract

The outbreak of COVID-19 Disease due to the novel coronavirus has caused a shortage of medical resources. To aid and accelerate the diagnosis process, automatic diagnosis of COVID-19 via deep learning models has recently been explored by researchers across the world. While different data-driven deep learning models have been developed to mitigate the diagnosis of COVID-19, the data itself is still scarce due to patient privacy concerns. Federated Learning (FL) is a natural solution because it allows different organizations to cooperatively learn an effective deep learning model without sharing raw data. However, recent studies show that FL still lacks privacy protection and may cause data leakage. We investigate this challenging problem by proposing a simple yet effective algorithm, named \textbf{F}ederated \textbf{L}earning \textbf{o}n Medical Datasets using \textbf{P}artial Networks (FLOP), that shares only a partial model between the server and clients. Extensive experiments on benchmark data and real-world healthcare tasks show that our approach achieves comparable or better performance while reducing the privacy and security risks. Of particular interest, we conduct experiments on the COVID-19 dataset and find that our FLOP algorithm can allow different hospitals to collaboratively and effectively train a partially shared model without sharing local patients' data.