A New Embedded Surveillance System for Reducing COVID-19 Outbreak in Elderly Based on Deep Learning and IoT

Abstract

As a result of the fast spread of Coronavirus disease (COVID-19) throughout the world, it became urgent to evolve an aided-intelligent system to help healthcare organizations to control and early detect COVID-19 outbreak, especially after massive development in the computing. Artificial Intelligence and Deep Learning methods could introduce real assistance to many healthcare organizations in this global problem by monitoring, detecting, and reporting on infected persons in early-stage. As is well known, the elderly are most vulnerable to the effects of COVID-19, therefore, we aim through this paper is to present the embedded system has the ability to detect and report on the elderly in the endemic areas. An age estimation from a facial image based on deep learning methods and Internet of Things is also applied to send notification to mobile or any other device systems that the application is embedded on it based on IoT. Depending on the experimental results on the proposed system through the Mean Average Error rate, the proposed system gave better or equivalent to the results of the state-of-the-art techniques, and the prediction results of an average accuracy achieved 89.45%. Finally, the proposed system has the capacity to help reduce the intensity of spread COVID-19 by identified older people and prevent them from prohibited in dangerous areas. © 2020 IEEE.