Why the study?
Imaging photoplethysmography enables simple, non-contact multiparameter measurement, but non-negative matrix factorization had not been evaluated against green channel traces for source signal acquisition.
Does non-negative matrix factorization improve the estimation of heart rate and respiratory rate compared to green channel trace in imaging photoplethysmography?
Does non-negative matrix factorization improve the estimation of heart rate and respiratory rate compared to green channel trace in imaging photoplethysmography?
The study explores non-negative matrix factorization as an alternative to green channel traces for extracting heart rate and respiratory rate from imaging photoplethysmography.
NMF may aid remote HR/RR monitoring via iPPG; leaves open superiority to green channel traces pending validation.
Imaging-Photoplethysmography has gathered a great amount of recognition in recent decades given to its simple, cost effective and non-contact multiparameter measurement capability. Generally, for blood volume pulse extraction, two processes are preferred. First, using green channel trace and second, using ICA or PCA to acquire underlying sources signal. ICA is used dominantly. Here, we have employed non-negative matrix factorization (NMF) for underlying source signal acquisition and compared it with the use of green channel for physiological parameter estimation. Normalized raw traces obtained from video recordings were pre-processed to keep only positive oscillations.
No takes yet. Share an insight, caveat, or question.
Singh et al. (2020) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: