Recent Advances in Non-Invasive Blood Pressure Monitoring and Prediction Using a Machine Learning Approach

Area of Science:

• Biomedical Engineering
• Health Informatics
• Wearable Technology

Background:

• Traditional blood pressure (BP) monitoring is often invasive or inconvenient for continuous use.
• Continuous non-invasive BP (NIBP) monitoring is crucial for proactive health assessment and disease management.
• Machine learning (ML) enhances the potential of NIBP for predictive health insights.

Purpose of the Study:

• To review the development of NIBP monitoring tools and sensors.
• To summarize state-of-the-art NIBP predictive techniques and ML applications.
• To compare ML models for NIBP measurement against international validation standards.

Main Methods:

• Review of sphygmomanometer development and NIBP sensors.
• Analysis of NIBP predictive techniques: pulse arrival time, pulse transit time, pulse wave velocity, and ML.
• Comparison of ML model performance against AAMI and BHS validation standards.

Main Results:

• Summary of current NIBP sensors and continuous monitoring approaches.
• Elaboration on ML-based NIBP prediction using bio-signals.
• Comparative analysis of ML models for systolic BP, diastolic BP, and mean arterial pressure.

Conclusions:

• Continuous NIBP monitoring holds significant potential for health assessment and early disease detection.
• ML integration is key to advancing NIBP accuracy and predictive capabilities.
• Addressing implementation challenges is vital for the widespread adoption of continuous NIBP technology.