Analysis of Instantaneous Linear, Nonlinear and Complex Cardiovascular Dynamics from Videophotoplethysmography

Summary

Background: There is a fast growing interest in the use of non-contact devices for health and performance assessment in humans. In particular, the use of non-contact videophotoplethysmography (vPPG) has been recently demonstrated as a feasible way to extract cardiovascular information. Nevertheless, proper validation of vPPG-derived heartbeat dynamics is still missing.

Objective: We aim to an in-depth validation of time-varying, linear and nonlinear/complex dynamics of the pulse rate variability extracted from vPPG.

Methods: We apply inhomogeneous pointprocess nonlinear models to assess instantaneous measures defined in the time, frequency, and bispectral domains as estimated through vPPG and standard ECG. Instantaneous complexity measures, such as the instantaneous Lyapunov exponents and the recently defined inhomogeneous point-process approximate and sample entropy, were estimated as well. Video recordings were processed using our recently proposed method based on zerophase principal component analysis. Experimental data were gathered from 60 young healthy subjects (age: 24±3 years) undergoing postural changes (rest-to-stand maneuver).

Results: Group averaged results show that there is an overall agreement between linear and nonlinear/complexity indices computed from ECG and vPPG during resting state conditions. However, important differences are found, particularly in the bispectral and complexity domains, in recordings where the subjects has been instructed to stand up.

Conclusions: Although significant differences exist between cardiovascular estimates from vPPG and ECG, it is very promising that instantaneous sympathovagal changes, as well as time-varying complex dynamics, were correctly identified, especially during resting state. In addition to a further improvement of the video signal quality, more research is advocated towards a more precise estimation of cardiovascular dynamics by a comprehensive nonlinear/complex paradigm specifically tailored to the non-contact quantification.

Keywords

Video-photoplethysmography - heart rate variability - point process - nonlinear analysis - complexity - bispectrum - sympatho-vagal balance - ECG

• References

• 1 Poh MZ, McDuff DJ, Picard RW. Advancements in noncontact, multiparameter physiological measurements using a webcam. IEEE Transactions on Biomedical Engineering 2011; 58 (01) 7-11.
• 2 Sun Y. et al. Noncontact imaging photoplethysmography to effectively access pulse rate variability. J Biomed Opt 2013; 18 (06) 061205.
• 3 de Haan G, Jeanne V. Robust pulse rate from chrominance-based rppg. IEEE Transactions on Biomedical Engineering 2013; 60 (10) 2878-2886.
• 4 Tarassenko L. et al. Non-contact video-based vital sign monitoring using ambient light and auto-regressive models. Physiological Measurement 2014; 35 (05) 807.
• 5 Takano C, Ohta Y. Heart rate measurement based on a time-lapse image. Medical Engineering & Physics 2007; 29 (08) 853-857.
• 6 Moreno J, Ramos-Castro J, Movellan J, Parrado E, Rodas G, Capdevila L. Facial video-based photoplethysmography to detect hrv at rest. Int J Sports Med 2015; 36 (06) 474-480.
• 7 Valenza G, Citi L, Scilingo EP, Barbieri R. Pointprocess nonlinear models with laguerre and volterra expansions: Instantaneous assessment of heartbeat dynamics. IEEE Transactions on Signal Processing 2013; 61 (11) 2914-2926.
• 8 Valenza G, Citi L, Scilingo EP, Barbieri R. Inhomogeneous point-process entropy: An instantaneous measure of complexity in discrete systems. Physical Review E 2014; 89 (05) 052803.
• 9 Valenza G, Citi L, Barbieri R. Estimation of instantaneous complex dynamics through Lyapunov exponents: a study on heartbeat dynamics. PloS One 2014; 09 (08) e105622.
• 10 Iozzia L, Cerina L, Mainardi L. Assessment of beat-to-beat heart rate detection method using a camera as contactless sensor. Conf Proc IEEE Eng Med Biol Soc 2016; 2016: 521-524.
• 11 Barbieri R, Matten EC, Alabi ARA, Brown EN. point-process A model of human heartbeat intervals: new definitions of heart rate and heart rate variability. American Journal of Physiology-Heart and Circulatory Physiology 2005; 288 (01) H424.
• 12 Fitzpatrick TB. The validity and practicality of sun-reactive skin types i through vi. Archives of Dermatology 1988; 124 (06) 869-871.
• 13 Viola P, Jones M. Rapid object detection using a boosted cascade of simple features. In: Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition 2001; I-511.
• 14 Tomasi C, Kanade T. Detection and tracking of point features. School of Computer Science. Carnegie Mellon Univ; Pittsburgh: 1991
• 15 Tarvainen MP, RantaAho PO, Karjalainen PA. Anadvanced detrending method with application to hrv analysis. IEEE Transactions on Biomedical Engineering 2002; 49 (02) 172-175.
• 16 Iozzia L, Cerina L, Mainardi L. Relationships between heart-rate variability and pulse-rate variability obtained from video-ppg signal using zca. Physiological Measurement 2016; 37 (11) 1934.
• 17 Afonso V, Tompkins WJ, Nguyen TQ, Luo S. ECG beat detection using filter banks. IEEE Transactions on Biomedical Engineering 1999; 46 (02) 192-202.
• 18 Valenza G, Citi L, Garcia RG, Noggle JTaylor, Toschi N, Barbieri R. atComplexity variability assessment of nonlinear time-varying cardiovascular control. Scientific Reports 2017; 07: 42779.