This study will focus on an in-depth survey of commercial and open source resources for face processing for telehealth using Remote Visual Photoplethysmography (rvPPG). We can use Remote Visual Photoplethysmography (rvPPG) as a tool to monitor vital health signs and use its results to observe health conditions. There is a growing demand for non- invasive remote (tele-services) patient care information technology solutions. Advancement in the area of machine learning and deep learning especially as it relates to computer vision made it possible to predict various vital signs such as heart rate, blood pressure and oxygen saturation from pixel changes in recorded or live video feeds of human skin. Ability to extract vital signs via face processing could be a major contribution for telehealth solutions. Recent studies have shown that rvPPG could be a sufficient substitute for traditional contact based methods for PPG, i.e. activity trackers, smart watches and pulse oximeter. In order to help the scientific community take advantage of the benefits of rvPPG methods, we would like to identify the rvPPG research and tools that currently exist in academia and the industry. There may be missed opportunities for tools developed for research purposes to be transferred to the industry. Our research will shed light on tools available therefore creating an opportunity for academia to discover what has been done in industry. The research will be unique in the sense that it will be an extensive survey of existing literature and tools. In this thesis, we will perform a systematic literature review of the rvPPG research and tools that exists today, and identify the various directions that this research area can take. This work will provide detailed information about open source datasets and that can be used for developing a rvPPG framework. Furthermore, we will compare the selected commercial and open source rvPPG tools’ specifications. In this review, we have included thirteen novel methods that have significant contributions, twelve open source public datasets, fifteen open source tools and eleven commercial tools for literature review. We also discovered that there is an opportunity for the transferal of rvPPG methods from academia to industry to aid face processing for telehealth software development and maintenance.

Recommended Citation: Demirhan H., Ricanek K., Simmonds D., Clark U., (2022). A Review of Face Processing for Telehealth: Research Survey of Remote Visual Photoplethysmography (rvPPG). UNCW MS CSIS Proceedings. V. 16 , N. 0 .