Paradigm Shift Towards Federated Learning for COVID-19 Detection: A Survey

Abstract

     The novel coronavirus 2019 (COVID-19) is a respiratory syndrome with similar traits to common pneumonia. This major pandemic has affected nations both socially and economically, disturbing everyday life and urging the scientific community to develop solutions for the diagnosis and prevention of COVID-19. Reverse transcriptase-polymerase chain reaction (RT–PCR) is the conventional approach used for detecting COVID-19. Nevertheless, the initial stage of the infection is less predictable in PCR tests, making early prediction challenging. A robust and alternative diagnostic method based on digital computerised technologies to support conventional methods would greatly help society. Therefore, this paper reviews recent research based on using machine and federated learning techniques on publicly available datasets comprising Computed Tomography (CT) images, Chest X-ray (CXR) and ultrasound of COVID-19 patients. This paper also analyses the analytical efficiency such as accuracy, sensitivity, specificity and F1-score of models to determine the efficacy. Based on our study, we observed that Machine Learning (ML) was proposed widely in COVID-19 prediction and diagnosis methods. But this method has challenges due to less dataset availability and privacy concerns. However, federated learning-based COVID-19 detection overcame the challenge and provided better efficacy with low datasets and supported medical data privacy. Thus, based on the advantage observed, federated learning-based COVID-19 detection systems should be developed in the future.