Cuffless, calibration-free hemodynamic monitoring with physics-informed machine learning models

Area of Science:

• Biomedical Engineering
• Cardiovascular Physiology
• Wearable Technology

Background:

• Current cuffless blood pressure (BP) monitoring devices lack theoretical foundations, leading to inaccuracies.
• Existing methods like pulse wave analysis are susceptible to various confounders.
• There is a need for reliable, continuous hemodynamic monitoring solutions.

Purpose of the Study:

• To develop a smartwatch with electrical bioimpedance (BioZ) sensing for cuffless hemodynamic monitoring.
• To establish the biophysical relationship between BioZ signals and BP.
• To enable calibration-free estimation of BP and blood velocity.

Main Methods:

• Developed a smartwatch with real-time electrical bioimpedance (BioZ) sensing.
• Utilized a multiscale analytical and computational modeling framework to link BioZ and BP.
• Employed a physics-informed neural network with fluid dynamics principles for BP and blood velocity estimation.

Main Results:

• Successfully demonstrated the feasibility of BioZ technology for cuffless monitoring.
• Identified key parameters influencing pulsatile BioZ signals at the wrist.
• Achieved calibration-free estimation of BP and blood velocity in diverse settings.

Conclusions:

• Electrical bioimpedance (BioZ) technology is a viable solution for cuffless hemodynamic monitoring.
• This approach addresses critical limitations of existing cuffless wearable devices.
• The developed smartwatch offers potential for transforming cardiovascular health assessments.