[Noninvasive Continuous Blood Pressure Measurement Method Based on EEMD and ANN]

Area of Science:

• Biomedical Engineering
• Cardiovascular Physiology
• Signal Processing

Background:

• Accurate blood pressure monitoring is crucial for cardiovascular health assessment.
• Current non-invasive methods often lack continuous measurement capabilities.
• Electronic sphygmomanometers require improved non-invasive continuous blood pressure monitoring.

Purpose of the Study:

• To propose a novel non-invasive method for continuous blood pressure measurement.
• To utilize ensemble empirical mode decomposition (EEMD) and artificial neural networks (ANN) for blood pressure estimation.
• To validate the accuracy of the proposed method against established standards.

Main Methods:

• Analysis of 19,500 pulse wave signals from the MIMIC DATABASE.
• Decomposition of pulse wave signals using EEMD.
• Extraction of 10 characteristic parameters from the 4th layer decomposition signal.
• Training an artificial neural network (ANN) model with extracted parameters to predict blood pressure.
• Error analysis of the trained blood pressure (BP) model.

Main Results:

• The EEMD technique effectively decomposed pulse wave signals.
• 10 key characteristic parameters were identified as effective inputs for the ANN.
• The trained ANN model demonstrated accurate blood pressure prediction.
• The model's error analysis met the stringent standards of the Association for the Advancement of Medical Instrumentation (AAMI).

Conclusions:

• The proposed EEMD-based ANN method offers a viable solution for non-invasive continuous blood pressure monitoring.
• This approach enhances the capabilities of electronic sphygmomanometers.
• The method's accuracy supports its potential clinical application in cardiovascular health management.