System for Evaluation of COVID-19 Prevention Guidelines Compliance

In a catastrophic medical situation caused by an infectious disease, such as COVID-19, it is very important to quickly determine who and where to be tested and supervised. The current COVID-19 screening test is conducted by identifying people with high probability of infection, such as who made direct or indirect contact with the confirmed person, by identifying the moving path of the confirmed person. Currently, various methods are being employed, such as interviewing or location tracking through cell phone forensics, to determine the moving path of the confirmed person. Mostly, however, these methods are time consuming, inaccurate, and easy to invade privacy while promptness, accuracy and anonymity are key values of epidemiological surveillance. There is still no preemptive management methods for a space where infection occurs are possible. Investigation and action on the area where the infection occurred are just carried out only after a confirmed person has been confirmed. In order to solve these problems, it is necessary to develop an automatic system for evaluating space for compliance of infectious disease prevention guidelines, or simply risk estimation system, using artificial intelligence and computer vision technology. In this paper, we discuss the system for evaluation of COVID-19 prevention guidelines compliance which has been researching and developing by ASSIST.