Video Photoplethysmography (vPPG) allows for estimation of blood volume pulse (BVP) from the skin by means of a video camera recording at high frequency rate. The estimation procedure presents several drawbacks in its application to real world conditions, such as light changes or movements that often generate artifacts in the extracted BVP waveform. In addition, the process requires a skin segmentation algorithm to distinguish skin pixels from the background. To date, even the most refined skin segmentation algorithms still need a manual definition that could lead to incorrect pixel classification, and consequently to a decrease in the signal-to-noise ratio (SNR). We here propose a fully autonomic procedure able to extract BVP from video recordings of the skin in real world conditions.The experimental protocol is designed to record the signals of interest and to evaluate changes in the Autonomic Nervous System modulation of the heart during a baseline condition and a controlled breathing phase. Video recordings are gathered from 4 young healthy subjects (age: 21±1 years). vPPG signals are processed in order to extract the BVP waveform, and a peak detection algorithm detects pulse wave peaks that are then used to compute specific measures of heart rate variability (HRV).The procedure is successfully validated by comparing the extracted HRV measures against those extracted using a finger photoplethysmograph (fPPG) using three different skin segmentation algorithms from BVP signals.

Assessing an Automatic Procedure of Extraction of Physiological Parameters from Skin using Video Photoplethysmography

Barbieri R.;Levi R.;Mollura M.;Marsella I.;Cerina L.;Mainardi L.
2020-01-01

Abstract

Video Photoplethysmography (vPPG) allows for estimation of blood volume pulse (BVP) from the skin by means of a video camera recording at high frequency rate. The estimation procedure presents several drawbacks in its application to real world conditions, such as light changes or movements that often generate artifacts in the extracted BVP waveform. In addition, the process requires a skin segmentation algorithm to distinguish skin pixels from the background. To date, even the most refined skin segmentation algorithms still need a manual definition that could lead to incorrect pixel classification, and consequently to a decrease in the signal-to-noise ratio (SNR). We here propose a fully autonomic procedure able to extract BVP from video recordings of the skin in real world conditions.The experimental protocol is designed to record the signals of interest and to evaluate changes in the Autonomic Nervous System modulation of the heart during a baseline condition and a controlled breathing phase. Video recordings are gathered from 4 young healthy subjects (age: 21±1 years). vPPG signals are processed in order to extract the BVP waveform, and a peak detection algorithm detects pulse wave peaks that are then used to compute specific measures of heart rate variability (HRV).The procedure is successfully validated by comparing the extracted HRV measures against those extracted using a finger photoplethysmograph (fPPG) using three different skin segmentation algorithms from BVP signals.
2020
Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS
978-1-7281-1990-8
Algorithms
Blood Volume
Heart Rate
Humans
Young Adult
Autonomic Nervous System
Photoplethysmography
File in questo prodotto:
File Dimensione Formato  
2020_01_Video_Photoplethysmography.pdf

accesso aperto

Descrizione: Articolo Principale
: Post-Print (DRAFT o Author’s Accepted Manuscript-AAM)
Dimensione 1.16 MB
Formato Adobe PDF
1.16 MB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11311/1170166
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 1
  • ???jsp.display-item.citation.isi??? 0
social impact