The Investigation Related to Application of Federated Learning’s in Chest X-ray Detection

Abstract

Following the global emergence of COVID-19 in 2020, the significance of Chest X-ray detection has grown exponentially as it plays a crucial role in diagnosing respiratory conditions. Concurrently, the evolving federated learning framework has been progressively integrated into the medical field, particularly in conjunction with Chest X-ray detection. This integration reflects a promising trend in enhancing collaborative diagnostic capabilities and leveraging collective knowledge across diverse medical institutions. Based on this background, this article provides a thorough review of medicine detection related federated learning frameworks and federated models, summarizes the characteristics and methods of federated models that have been widely used in various experiments in recent years, discusses and analyzes their advantages and disadvantages, and compares their performance with existing other machine learning models. In conclusion, the federated model outperforms non-federated machine learning models when it comes to analyzing Chest X-ray images and predicting symptoms. Lastly, this article outlines potential risks and offers improvement suggestions for the implementation of federated learning in chest X-ray detection.