Pseudo-Data based Self-Supervised Federated Learning for Classification of Histopathological Images

TL;DR

This paper introduces a novel federated learning framework using pseudo-data and self-supervised learning to enhance diagnostic accuracy and generalization of histopathological image classification models across multiple centers.

Contribution

It proposes a pseudo-data based self-supervised federated learning framework with a multi-task SSL and a Barlow Twins based contrastive learning algorithm to improve model performance and privacy.

Findings

Improved diagnostic accuracy on three public datasets.

Enhanced model generalization across different centers.

Effective privacy-preserving pseudo-data sharing.

Abstract

Computer-aided diagnosis (CAD) can help pathologists improve diagnostic accuracy together with consistency and repeatability for cancers. However, the CAD models trained with the histopathological images only from a single center (hospital) generally suffer from the generalization problem due to the straining inconsistencies among different centers. In this work, we propose a pseudo-data based self-supervised federated learning (FL) framework, named SSL-FT-BT, to improve both the diagnostic accuracy and generalization of CAD models. Specifically, the pseudo histopathological images are generated from each center, which contains inherent and specific properties corresponding to the real images in this center, but does not include the privacy information. These pseudo images are then shared in the central server for self-supervised learning (SSL). A multi-task SSL is then designed to fully learn both the center-specific information and common inherent representation according to the data characteristics. Moreover, a novel Barlow Twins based FL (FL-BT) algorithm is proposed to improve the local training for the CAD model in each center by conducting contrastive learning, which benefits the optimization of the global model in the FL procedure. The experimental results on three public histopathological image datasets indicate the effectiveness of the proposed SSL-FL-BT on both diagnostic accuracy and generalization.