Spatiotemporal processing in photoplethysmography for skin microcirculatory perfusion imaging

Technological advances in the real-time visualization of cutaneous microcirculation aim to realize benefits including high-resolution imaging, suppressed noise, and robust temporal coherence. Photoplethysmography (PPG), a noninvasive technique that measures single or multiple points of relative blood volume changes in blood vessels under the skin, shows potential as a signal candidate for visualizing blood vessels and tracking blood flow. However, challenges still remain, such as extracting/image reconstruction of the blood vessel/flow signal in a precise frequency window (<0.2 Hz) from a noisy image that is caused by the loss of spatial coherence of the light source in a turbid biological tissue. We attempted to overcome this challenge by adopting a combination of direct-contact-type, lens-less, conformable imagers and singular value decomposition (SVD) in this study. We focused on the numerical analysis of SVD for discriminating the tissue and vein blood flow in PPG for reconstructing blood fluidic images, followed by a complete demonstration of skin microcirculation blood tracking in the vessel visualization process when applying our lens-less, conformable, wearable imagers.

Automated summary

This study focuses on overcoming challenges in processing photoplethysmography (PPG) signals by using a combination of direct-contact-type, lens-less, conformable imagers and singular value decomposition (SVD) technology. The research successfully demonstrated skin microcirculation blood tracking and vessel visualization by applying the lens-less, conformable, wearable imagers.