Blending of Internet of Things and Deep Transfer Learning (DTL): Enabling Innovations in Healthcare (COVID-19) and Applications

Abstract

Global healthcare systems are currently facing unprecedented hardships while handling the pandemic arising out of the COVID-19 virus. The risks of widespread fatality, inadequate medical infrastructure, and the infections arising out of the COVID-19 virus have necessitated a wide range of requirements to formulate appropriate damage minimization response. It provides the opportunity for extensive use of technology. Among many such technologies, the Internet of things (IoT) has emerged as a viable one as it combines a host of sensor packs, wireless communication-based networking, and automated decision making when combined with artificial intelligence (AI) tools. Further, emerging applications involving edge computing (EC), deep learning (DL) and Deep Transfer Learning (DTL) have enabled IoT to evolve and act decisively in the healthcare sector and extend helping hands for lowering of risks in the midst of pandemic situations. In this chapter, we present a composite framework formulated using IoT, DTL, and EC techniques for monitoring COVID-19 cases. These approaches could be used to alleviate risk-filled situations including those linked with the spread and fatalities due to COVID-19. The proposed approach is suitable for monitoring the spread of deadly virus, identifying this infectious disease, discovering booster dose, discovering drugs, and related aspects. We additionally point to certain cases where DL can help in providing effective prediction of the spread, disinformation identification, and related opinion analysis, although DL requires billions datasets and different computing tools. As DTL takes data from one task and it could work on the other, DTL techniques would be more effective than DL. Edge devices (ED) are also important in this pandemic situation as sophisticated infrastructure is essential to face and handle pandemic-like situation by providing complementary-supplementary support. It is essential for risk minimization especially as a support to the frontline workers who face the COVID-19 infected patients directly. EDs are automated and also may help in fighting crises arising out of this outbreak. Here, a comparative analysis is reported which deals with DL and other learning based tools, which are specially configured along with IoT setups for dealing with medical emergencies and pandemic situations arising out of COVID-19. We highlight some of the technologies and the applications used in the pandemic situation.