Photoplethysmography Imaging Algorithm for Real-Time Monitoring of Skin Perfusion Maps

Uldis Rubins, Aleksejs Miscuks, Yousef Qawqzeh, Zbignevs Marcinkevics, Andris Grabovskis; Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2023, pp. 5950-5956

Photoplethysmography imaging (PPGI) is a cost-effective and reliable contactless optical technique for assessing cutaneous microcirculation. Its applications extend beyond extracting vital signs, such as heart rate and respiratory rate, to include spatiotemporal perfusion measurements from human skin. We propose an algorithm that provides real-time estimation of perfusion maps with adequate spatial resolution (320x240 pixels) and temporal resolution (0.2 s), which can be executed on a mid-range laptop computer. The algorithm was tested for determining perfusion changes in skin regions during regional anesthesia (RA) procedures. Seven patients (aged 18-80 years) undergoing hand surgery received peripheral nerve brachial plexus blocks during RA procedures. At baseline and after the RA procedure, 4, 10, 15, and 20 minutes later, the perfusion map density showed median values of 53.7%, 57.3%, 64.3%, 70.6%, and 71.1%, respectively, indicating an increase in perfusion due to the local anesthetic. Our easy-to-implement real-time video processing algorithm demonstrates potential for skin perfusion monitoring during RA procedures.