Graph connection Laplacian allows for enhanced outcomes of consumer camera based photoplethysmography imaging

Abstract

AbstractObjectThis work introduces a novel method to minimize the effect of global phase deviation that is inherent in photoplethysmographic images (PPGI) captured by video.MethodWe analyzed the facial vascular network obtained from a consumer camera as a two-dimensional manifold. Subtle phase variations across skin sites are due to complex dynamics of the vascular tree. Utilizing PPGI, the phase is modeled as a vector field of the facial manifold. The phase variations over different skin sites are caused by different blood volume modulations. We propose using the Graph Connection Laplacian (GCL) technique to quantify the global phase deviation, with the hope that correcting for this deviation could improve the quality of the PPGI signal. It is also postulated that study of this phase deviation might reveal valuable anatomical and physiological information.ResultThe proposed algorithm appears to yield a higher-quality global PPGI signal. By correcting the global phase deviation estimated by GCL waveform features such as the dicrotic notch are emphasized. The perfusion map, with the global phase deviation (estimated by GCL as intensity), appears to reflect skin perfusion dynamics.ConclusionThis algorithm enhances the quality of the global PPGI signal, facilitating the analysis of morphological parameters and showing promise for advancing PPGI applications in scientific research and clinical practice.