Federated Learning-based Healthcare Services for COVID-19 Pandemic

Healthcare data needs to maintain privacy and security, as it holds sensitive information of patients which can lead to identity theft. In case of centralized data sharing, all the features of data can be seen from the centralized server, thus exploitation of sensitive data can take place. By incorporating decentralized data sharing such as Federated Learning (FL), data privacy and security can be improved as the each component in the FL system contains only parts of data. Privacy will be achieved successfully without any data leakage. This preserves data sovereignty, and could reduce cyber-attacks. Thus, this survey discuses about the COVID-19 prediction using learning techniques by embracing FL in healthcare sector to addresses privacy, security and scalability concerns.