E M O T Y . A I

Deviceless Neurotechnologies

The photoplethysmography (PPG) which means heart-rate pulsation is generally measured on a finger or an ear using contact sensors. The recent several studies using web-cam to measure PPG have been introduced under the desktop or mobile computing environment including service robots.

It provides more comfortable measurement in or out of the clinic with the rapidly developing non-contact PPG measurement. People can watch their health values from smartphones. by integrating remote ppg technology without attaching any sensor. Now let’s get into a little detail of this wonderful technology without boring you.

 If we begin with the working principle of standard ppg, we can explain it by figure (1) easily.

As you can see from the figure 1; The rays emitted from the light source are absorbed by the skin and the rays that are not absorbed are reflected back and captured by the detector.

The relationship between the amount of light sent at the beginning and the amount of light captured gives the amount of blood volume in the vein.

PPG to rPPG

It was the normal PPG measurement which used contact sensors but there is the same logic in the contactless measurement. The heart rate depends on the autonomic nervous system and does not occur under the control of the person.

In this way, the heart rate can be measured from a computer or a phone camera without using any sensor based on the color intensity formed on the face. Remote ppg provides great convenience in the analysis of emotional states such as fear, excitement, stress, happiness.

Figure 2:
The setting of remote heart rate measurement.

A single 1-D signal is extracted from the spatially averaged values of these pixels over time. In parallel, 3-D head movements are tracked and used to suppress motion noise. An FFT based wide and narrow band filtering process produces a clean pulse waveform from which peaks are detected. The inter-beat intervals obtained from these peaks are then used to compute heart rate and heart rate variability. The full analysis can be performed in real time on a CPU.

Figure 3
An overview of the proposed heart rate and heart rate variability estimation pipeline (left to right). The face in captured webcam images are detected and modelled to track the skin pixels in region of interest.

The general approach to remote heart rate measurement

1.Skin pixel selection: First, it should be adapted to face. Blood intensity relative to time in the interest area of the face is the first and basic step to measure heart rate remotely. This region is called the Region of Interested Area (ROI).

2.Signal extraction: Each pixel color of the region (red, green, blue) average is measured over time.

3.Signal filtering: Noise from head movements to face model are detected by observing and then noiseless heart rate is produced.

4.Output calculations: Peak intervals are measured by detecting peaks, and then heart rate and heart rate variability are estimated.

Although the methods or algorithms used according to different studies vary, the steps generally applied are as mentioned. With the combination of appropriate programming languages (mostly MatLab and C) and algorithms, a suitable application is developed for non-contact heart rate measurement.

 References

https://arxiv.org/abs/1909.01206

https://pubmed.ncbi.nlm.nih.gov/26737108/

https://pubmed.ncbi.nlm.nih.gov/23366353/

https://ieeexplore.ieee.org/document/7358857

Leave a Comment

en_USEnglish
ko_KRKorean en_USEnglish