Beat-to-Beat Oscillometric Blood Pressure Estimation: A Bayesian Approach With System Identification

Area of Science:

• Biomedical Engineering
• Cardiovascular Physiology
• Medical Devices

Background:

• Current noninvasive blood pressure (BP) monitoring devices have limitations in providing continuous, beat-to-beat accuracy.
• Physiologically based mathematical models are crucial for advancing BP estimation techniques.

Purpose of the Study:

• To develop and validate innovative beat-to-beat oscillometric methods for noninvasive BP estimation.
• To overcome limitations of existing devices by providing continuous and accurate BP readings.
• To utilize physiologically grounded mathematical models for enhanced BP monitoring.

Main Methods:

• Developed novel beat-to-beat oscillometric BP estimation techniques.
• Employed physiologically based mathematical models of intra-arterial BP and arterial system dynamics.
• Utilized a recursive Bayesian method for parameter estimation and a system identification technique for refinement.
• Validated methods through simulations and experiments with 10 human subjects.

Main Results:

• Achieved mean errors as low as -1.26 mmHg for systolic BP and 2.03 mmHg for diastolic BP.
• Reported standard deviations of errors at 5.95 mmHg (systolic) and 4.16 mmHg (diastolic).
• Enabled estimation of additional cardiovascular parameters including heart rate, respiration rate, and mean arterial pressure.

Conclusions:

• The novel beat-to-beat oscillometric methods represent a significant advancement in noninvasive BP monitoring.
• These methods address current device limitations by offering continuous, beat-to-beat BP estimates.
• The approach shows promise for improving the reliability and comprehensiveness of cardiovascular monitoring devices.