Open AccessNon-contact measurement of oxygen saturation with an RGB camera
Author(s) -
Alessandro Guazzi,
Mauricio Villarroel,
João Jorge,
Jonathan Daly,
Matthew C. Frise,
Peter A. Robbins,
Lionel Tarassenko
Publication year - 2015
Publication title -
biomedical optics express
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.362
H-Index - 86
ISSN - 2156-7085
DOI - 10.1364/boe.6.003320
Subject(s) - rgb color model , oxygen saturation , region of interest , computer vision , saturation (graph theory) , oxygen , artificial intelligence , optics , track (disk drive) , computer science , materials science , biomedical engineering , physics , mathematics , medicine , quantum mechanics , combinatorics , operating system
A novel method (Sophia) is presented to track oxygen saturation changes in a controlled environment using an RGB camera placed approximately 1.5 m away from the subject. The method is evaluated on five healthy volunteers (Fitzpatrick skin phenotypes II, III, and IV) whose oxygen saturations were varied between 80% and 100% in a purpose-built chamber over 40 minutes each. The method carefully selects regions of interest (ROI) in the camera image by calculating signal-to-noise ratios for each ROI. This allows it to track changes in oxygen saturation accurately with respect to a conventional pulse oximeter (median coefficient of determination, 0.85).