CoViS: A Contactless Health Monitoring System for the Nursing Home
Sérgio Ivan Lopes, Fábio Silva, Pedro Pinho, Paulo Marques, Carlos Abreu, João Milheiro, Bruno P. Braga, Gabriel Queirós, Rita de Cássia Almeida, Nuno Borges Carvalho
Abstract
In a pandemic, the availability of health indicators among at-risk populations, such as the elderly, is vital due to the disease’s rapid transmission and the need to act quickly to contain its evolution. Traditional clinical methods for monitoring vital signs typically require contact-based sensors that must be precisely attached by a skilled healthcare professional, are less practical for repeatable measurements, and are not suitable for long-term monitoring. Contactless vital signs monitoring, such as radar-based techniques or IR-thermal imaging, on the other hand, does not require the attachment of physical electrodes and can be useful for remote health monitoring and therefore reduce physical contact between subjects and healthcare professionals, ensuring social distancing, in particular, in nursing homes, and protecting the elderly population in this way from other external factors that increase the likelihood of infection. This article presents three main contributions: (i) vital signs characterization in the elderly population; (ii) a state-of-the-art review on the most prominent techniques and methods for Contactless Health Monitoring (CHM); (iii) the design, specification, and evaluation of a low-cost proof-of-concept CHM system for nursing homes, which includes an IoT Edge Device , which enables real-time vital signs monitoring (cardio-respiratory rates and elevated body temperature) using a multimodal approach based on Doppler radar and IR thermal imaging sensors, and thus generates health indicators without any type of contact or invasiveness. By direct comparison with reference instruments, results have shown an error below 10%, in 74%, 52%, and 96% of the cases for Heart Rate, Respiratory Rate, and Body Temperature measurements, respectively.