Collaborative Federated Learning for Healthcare: Multi-Modal COVID-19 Diagnosis at the Edge

Abstract

Despite significant improvements over the last few years, cloud-based healthcare applications continue to suffer from poor adoption due to their limitations in meeting stringent security, privacy, and quality of service requirements (such as low latency). The edge computing trend, along with techniques for distributed machine learning such as federated learning, has gained popularity as a viable solution in such settings. In this paper, we leverage the capabilities of edge computing in medicine by evaluating the potential of intelligent processing of clinical data at the edge. We utilized the emerging concept of clustered federated learning (CFL) for an automatic COVID-19 diagnosis. We evaluate the performance of the proposed framework under different experimental setups on two benchmark datasets. Promising results are obtained on both datasets resulting in comparable results against the central baseline where the specialized models (i.e., each on a specific image modality) are trained with central data, and improvements of 16% and 11% in overall F1-Scores have been achieved over the trained model trained (using multi-modal COVID-19 data) in the CFL setup on X-ray and Ultrasound datasets, respectively. We also discussed the associated challenges, technologies, and techniques available for deploying ML at the edge in such privacy and delay-sensitive applications.