Insights on Using Time-of-Flight Camera for Recovering Cardiac Pulse From Chest Motion in Depth Videos.

Area of Science:

• Biomedical Engineering
• Medical Imaging
• Physiological Monitoring

Background:

• Extracting cardiac pulse signals non-invasively is crucial for physiological monitoring.
• Existing methods may raise privacy concerns or require physical contact.
• Depth imaging offers potential for contactless physiological measurements.

Purpose of the Study:

• To introduce a novel method for extracting cardiac pulse signals from the human chest using depth camera technology.
• To demonstrate the ability to isolate subtle heartbeat motions from chest movements.
• To validate the accuracy and robustness of the developed technique.

Main Methods:

• Utilized a depth camera with a near-infrared sensor and time-of-flight technology to capture depth images.
• Developed coherent processing techniques to isolate cardiac pulse signals by suppressing interference from breathing and body motion.
• Applied micromotion reconstruction using depth image intensity for heartbeat analysis.

Main Results:

• Successfully extracted clean cardiac pulse signals from the chest area, demonstrating high consistency with contact photoplethysmography for inter-beat-interval derivation.
• Developed a graphical interpretation of pulsatile components in eigen space.
• Validated the method on ten healthy subjects, showing consistent results across different resting heart rates.

Conclusions:

• The novel depth camera approach provides a privacy-preserving, contactless method for cardiac pulse extraction.
• The developed signal processing techniques effectively mitigate motion artifacts, enabling accurate physiological measurements.
• The method shows robustness across various conditions, including different clothing thicknesses and measurement sites.