Pulse rate variability analysis by video using face detection and tracking algorithms

Heart rate variability (HRV) is a physiological parameter which has increased its importance in recent years due to its relationship with the autonomic nervous system (ANS) and cardiovascular disorders. In this work the pulse rate variability (PRV) is measured as a surrogate measurement of HRV using only a video of the face of the subject. Because of pulse rate signals obtained by video are largely affected by the person movements and the analyzed region of interest (ROI), we propose the application of face detection and tracking algorithms as a solution to the artifacts produced by these factors. We compare the obtained results of the proposed method with a pulse sensor signal of a reference system. The statistical analysis results show good agreement between both methods.