Continuous Blood Pressure Monitoring Algorithm using Laser Doppler Flowmetry.

Area of Science:

• Biomedical Engineering
• Medical Devices
• Machine Learning in Healthcare

Background:

• Continuous blood pressure monitoring is crucial for managing cardiovascular health.
• Existing cuff-based methods can be obtrusive and interrupt daily activities.
• Noninvasive, continuous monitoring techniques are highly sought after.

Purpose of the Study:

• To introduce a novel machine learning algorithm for continuous blood pressure estimation.
• To utilize Laser Doppler Flowmetry (LDF) for noninvasive blood flow measurements.
• To validate the algorithm's performance against established clinical standards.

Main Methods:

• Developing a machine learning algorithm to process LDF signals.
• Segmenting blood flow profiles based on heartbeat cycles.
• Extracting relevant features for blood pressure estimation.
• Employing a multi-layer neural network for beat-to-beat analysis.
• Validating results with cuff-based continuous blood pressure sensors.

Main Results:

• The algorithm successfully estimated beat-to-beat blood pressure from LDF data.
• Mean average error values ranged from 4.54 to 5.37 mmHg.
• Performance met Grade B/C criteria according to IEEE standard 1708-2014 for cuffless devices.

Conclusions:

• The proposed machine learning algorithm offers a viable noninvasive method for continuous blood pressure monitoring.
• LDF combined with advanced algorithms shows potential for accurate and convenient blood pressure assessment.
• This technology could significantly improve patient care and cardiovascular health management.