Estimating blood pressure using Windkessel model on Photoplethysmogram

Area of Science:

• Biomedical Engineering
• Medical Devices
• Physiological Monitoring

Background:

• Non-invasive vital sign estimation is crucial for preventive healthcare.
• Photoplethysmography (PPG) is a widely used non-invasive optical technique to detect blood volume changes.
• Accurate blood pressure (BP) monitoring remains a challenge, especially with non-invasive methods.

Purpose of the Study:

• To develop and validate a novel methodology for estimating blood pressure (BP) using photoplethysmography (PPG) signals.
• To improve the accuracy and robustness of non-invasive BP estimation compared to existing methods.
• To explore an indirect approach for BP estimation by mapping PPG features to intermediate physiological parameters.

Main Methods:

• Utilized PPG signals to estimate person-specific intermediate latent parameters, including total peripheral resistance and arterial compliance, via a 2-Element Windkessel model.
• Employed linear regression to derive systolic and diastolic blood pressure (BP) values from these estimated latent parameters.
• Validated the methodology on a standard hospital dataset.

Main Results:

• Achieved low absolute errors for BP estimation: 0.78±13.1 mmHg for systolic BP and 0.59 ± 10.23 mmHg for diastolic BP.
• Demonstrated that the proposed methodology is more robust than standard methods that directly estimate BP from PPG features.
• The indirect mapping approach showed improved performance in BP estimation.

Conclusions:

• The proposed methodology provides a robust and accurate non-invasive method for estimating blood pressure using PPG.
• This indirect estimation approach, leveraging physiological models, offers a promising advancement in wearable health technology and preventive healthcare.
• Further research can explore refining the intermediate parameter estimation and expanding the validation dataset.