Blood pressure prediction based on multi-sensor information fusion of electrocardiogram, photoplethysmography, and pressure pulse waveform

Cardiovascular diseases are now the leading cause of death that endangers people's health. Thus, a precise and dependable blood pressure (BP) prediction method is essential. This paper proposes a noninvasive BP prediction method with the fusion of electrocardiogram (ECG), photoplethysmography (PPG), and pressure pulse waveform (PPW) features. A multi-sensor information acquisition platform was developed. Besides, the algorithms were designed to clean, preprocess, and extract features from sample data. Finally, feature selection and feature fusion created the Random Forest Regression (RFR) BP prediction model feature set. The RFR predicted diastolic blood pressure with a mean absolute error (MAE) of 0.90 mmHg, a mean square error (MSE) of 6.48 mmHg, and a standard deviation (STD) of 2.47 mmHg; and systolic blood pressure with a mean absolute error (MAE) of 0.64 mmHg, a mean square error (MSE) of 2.31 mmHg, and a standard deviation (STD) of 1.53 mmHg.