CXR-FL: Deep Learning-Based Chest X-ray Image Analysis Using Federated Learning

TL;DR

This paper evaluates federated learning for deep learning-based chest X-ray analysis, highlighting its impact on model performance, interpretability, and generalizability while preserving data privacy across multiple centers.

Contribution

It provides an empirical assessment of federated learning parameters and compares full image versus lung segmentation training approaches for chest X-ray classification.

Findings

Federated learning maintains model generalizability across centers.

Training on lung regions improves pathology interpretability.

Full images yield better classification performance.

Abstract

Federated learning enables building a shared model from multicentre data while storing the training data locally for privacy. In this paper, we present an evaluation (called CXR-FL) of deep learning-based models for chest X-ray image analysis using the federated learning method. We examine the impact of federated learning parameters on the performance of central models. Additionally, we show that classification models perform worse if trained on a region of interest reduced to segmentation of the lung compared to the full image. However, focusing training of the classification model on the lung area may result in improved pathology interpretability during inference. We also find that federated learning helps maintain model generalizability. The pre-trained weights and code are publicly available at (https://github.com/SanoScience/CXR-FL).