Use of machine learning in diagnosing COVID-19 infection

The world has witnessed the most devastating pandemic due to the rapid spread of COVID-19, an infectious disease caused by severe acute respiratory syndrome coronavirus (SARS-CoV2 virus). The public health emergency of international concern arose due to the sudden outbreak of COVID-19 where both medical and socio-economic structures remain entirely altered not only in developed countries but also in developing countries. In this crucial scenario, advanced technologies like machine learning (ML) and deep learning (DL) assisted the researchers and helped governments and other health officials (including frontline workers) to manage the outbreak. ML is a sub-branch of computer science, where, machines can analyze large datasets and derive inference from that variable data structures. With the help of suitable algorithms, computers can imitate human behavior by analyzing results and the machines can perform in less time with great accuracy. During the pandemic, due to the scarcity of human resources, ML aided in the diagnosis of patients, forecasted communal transmission, and also helped in the development of effective antivirals and vaccines. In this chapter, we have highlighted the importance of various state-of-the-art ML tools, algorithms and computational models useful in the diagnosis and management of COVID-19. The circumstantial applications of ML are also discussed with real-time case studies. Lastly, the challenges faced by ML in COVID-19 supervision and future directions are also discussed. This chapter will help the researchers and students to understand how this powerful tool is employed to fight COVID-19 and can assist in future health emergencies due to emerging pathogens.