Pulse rate variability estimation method based on imaging-photoplethysmography and application to telepsychiatry

Area of Science:

• Biomedical Engineering
• Computer Vision
• Physiological Monitoring

Background:

• Telehealth adoption is increasing, offering wider access to mental health services.
• Non-contact physiological monitoring is crucial for remote patient assessment.
• Current methods for remote vital sign monitoring can be limited.

Purpose of the Study:

• To develop and validate a novel algorithm for extracting pulse rate (PR) and respiration rate (RR) signals from non-contact facial image sequences.
• To assess the accuracy of the algorithm in monitoring PR, RR, and pulse rate variability (PRV) under resting conditions.
• To explore the potential clinical relevance of this technology in telepsychiatry.

Main Methods:

• Image processing techniques for skin pixel extraction from facial images.
• Signal reconstruction utilizing spectral information from RGB signals.
• Algorithm validation on 15 healthy subjects in a controlled laboratory setting.

Main Results:

• Accurate monitoring of respiration rate (RR) was achieved.
• Accurate monitoring of pulse rate (PR) was achieved.
• Accurate monitoring of pulse rate variability (PRV) was achieved under rest conditions.

Conclusions:

• The developed algorithm enables reliable, non-contact extraction of PR and RR signals from facial image sequences.
• This technology holds significant potential for supporting psychiatrists in telepsychiatry.
• Non-contact vital sign monitoring via facial analysis can improve remote mental healthcare accessibility.