Noninvasive Cuffless Blood Pressure Estimation With Dendritic Neural Regression

Area of Science:

• Biomedical Engineering
• Artificial Intelligence in Healthcare
• Cardiovascular Monitoring

Background:

• Blood pressure (BP) is a critical health indicator, but traditional cuff-based methods are inconvenient for continuous monitoring.
• Noninvasive cuffless BP measurement is a growing area of research due to the limitations of existing techniques.
• Machine learning (ML) algorithms using photoplethysmogram (PPG) and electrocardiogram (ECG) signals are common for BP estimation.

Purpose of the Study:

• To investigate the performance of a novel dendritic neural regression (DNR) method for blood pressure estimation.
• To evaluate DNR's ability to capture nonlinear relationships between ECG and PPG signal features for improved BP accuracy.
• To compare DNR's BP estimation capabilities against state-of-the-art ML techniques.

Main Methods:

• Developed a novel dendritic neural regression (DNR) model utilizing multiplication operators in dendritic branches for enhanced nonlinear feature capture.
• Employed the AMSGrad optimization algorithm to further improve the DNR model's performance.
• Integrated raw features from ECG and PPG signals with components of BP mathematical models as input for the DNR model.

Main Results:

• The DNR method significantly increased the accuracy of systolic BP, diastolic BP, and mean arterial pressure estimation.
• DNR performance was found to be superior to existing state-of-the-art machine learning techniques.
• The proposed method achieved a Grade A for long-term BP estimation according to the British Hypertension Society protocol.

Conclusions:

• The novel dendritic neural regression (DNR) method offers a reliable and accurate approach for noninvasive, cuffless, continuous blood pressure monitoring.
• DNR's architectural simplicity and powerful performance make it a promising tool for long-term BP assessment.
• This method addresses the limitations of traditional BP measurement, paving the way for improved cardiovascular health management.