[20a-A25-1~10] 3.3 Biomedical optics

9:00 AM - 9:15 AM JST(12:00 AM - 12:15 AM UTC)English Presentation

[20a-A25-1] A novel proposal to obtain cardiovascular parameters from remote photoplethysmography

〇(DC)Sarai Dominguez Hernandez¹, Gonzalo Paez¹ (1.Centro de Investigaciones en Optica)

Keywords:

remote photoplethysmography,signal-to-noise ratio,cardiovascular parameters

The COVID-19 pandemic has increased the demand for telemedicine services due to the risk of infection in physical hospitals. The services offered by telemedicine include heart rate (HR) and heart rate variability (HRV). HR is a key physiological sign that acts as an indicator of a person's total cardiac output and as a preventive clinical diagnostic tool; especially important for early detection of cardiac arrhythmias and cardiovascular diseases by monitoring health parameters. Remote photoplethysmography (rPPG) is a non-contact method of measuring cardiovascular health parameters. It is very useful in cases of high risk of infection, such as the COVID-19 pandemic. Broadly, rPPG uses a digital camera (working as a photodetector) to capture subtle changes in the skin of the subject's face on video without the requirement for a special light source. After image processing, it is possible to extract the blood flow information and estimate HR. This method offers an alternative to monitor HR because they do not require dedicated wearable devices or electrode pads required for photoplethysmogram (PPGm) and electrocardiogram measurements. It is critical to obtain a high-quality rPPG signal to accurately measure cardiovascular health parameters. This has prompted the research and development of advanced signal-processing techniques involving the improvement of the signal-to-noise ratio, which is one of the main challenges (Haugg 2023). We propose novel and robust techniques to accurately calculate heart rate using rPPG. We use the Gaussian filter, Kalman filter and adaptative filter to improve the signal-to-noise ratio, which allows us to calculate the heart rate and extract information of physiological interest. Our proposal provides a solution to the noise problem that the method itself implies, allowing us to focus the efforts of rPPG research on extracting relevant physiological information from it, instead of looking for alternatives to reduce the noise.