Advancements in Federated Learning for Health Applications: A Concise Survey
Vasileios Stamatis, Panagiotis I. Radoglou Grammatikis, Antonios G. Sarigiannidis, Nikolaos Pitropakis, Θωμάς Λάγκας, Vasileios Argyriou, Evangelos Markakis, Panagiotis G. Sarigiannidis · 2024
Smart solutions in the healthcare domain have garnered considerable attention due to their potential to enhance standard treatment methods and improve overall health. However, privacy concerns often prevent the sharing of healthcare data, which can limit the scope for improvement. In this context, Federated Learning (FL) has emerged as a transformative paradigm in machine learning. It enables collaborative model training across decentralised devices while preserving data privacy and security. This approach has gained significant traction in recent years, particularly within the healthcare sector. It offers unprecedented opportunities to harness collective intelligence from diverse healthcare datasets without compromising sensitive patient information. This survey paper summarises numerous research works that focus on the application of FL to address various healthcare challenges. Moreover, a comparison of these works is conducted, summarising the different technologies employed in each case. Therefore, in light of the previous remarks, this paper provides an up-to-date overview of the state of the art in the application of FL in the healthcare industry.