Internet of Things (IoT) Model for the Detection of an Infectious Disease (COVID-19)

Abstract

The COVID-19 pandemic, originating in late 2019 due to the highly transmissible SARS-CoV-2 virus, has precipitated a global health crisis with profound impacts on healthcare systems, economies, and societal structures. Despite advancements in vaccination and treatment development, persistent challenges endure due to viral mutations, necessitating continuous vigilance and robust screening efforts. In response, remote photoplethysmography (rPPG) technology has emerged as a critical tool for contactless heart monitoring during COVID-19 screening protocols. This innovation reduces virus transmission risks by eliminating physical contact during vital sign assessments, capturing crucial data including heart rate, body temperature, and oxygensaturation levels. The presented thesis investigates the utilization of IoT devices, incorporating an RGB camera and an infrared camera, to non invasively predict the presence of COVID-19. The methodology entails video capture, frame extraction, facial detection techniques, and prediction of vital signs including body temperature, heart rate, and oxygen saturation. Leveraging an artificial neural network trained on a COVID-19 dataset, the implemented system achieves an impressive 95% accuracy in infection prediction. This system offers promising prospects to mitigate infection risks, enhance case detection, and find application across various settings,
including entry points, containment zones, and home quarantine