Machine Learning Solutions in Combating COVID-19: State of the Art and Challenges

Abstract

COVID-19 is a toxic virus that emerged in China and caused an epidemic globally. COVID-19 virus patients are placed in isolation so that the virus does not blow out widely. The only approach to secure people from this deadly virus is by maintaining social distance among people, wearing gloves and masks, and sanitizing and washing hands regularly. The law enforcement agencies and Government are included in prohibiting people movement in varied cities to control the virus spread. It is not feasible for government to supervise all places namely hospitals, shopping malls, banks, government offices and direct people to following the safety guidelines prescribed by government. The COVID-19 virus has been spreading in a drastic way with the feasibility of a restricted amount of testing kits rapidly. Therefore, the diagnosis of COVID-19 model is important to recognize the occurrence of disease from radiological images. The major aim of the research is to review different machine learning solutions in combating COVID-19. Various machine learning models are implemented to combat COVID-19 virus. The proposed models with the along computer vision algorithm are used as a corresponding device to be fixed at varied places and supervise people to adopt secure guidelines which are suggested by the government. By following the precautionary steps and using the solutions of machine learning, people are capable to succeed with the fight against COVID-19 virus. Keywords: COVID-19, machine learning, social distancing, face mask, hand hygiene, deep learning