1-Dimensional Convolutional Neural Network Based Blood Pressure Estimation with Photo plethysmography Signals and Semi-Classical Signal Analysis

In this paper, we propose a 1-Dimensional Convolutional Neural Network (1D-CNN) based Blood Pressure (BP) estimation using Photo plethysmography (PPG) signals and their features obtained through Semi-classical Signal Analysis (SCSA). The procedure of the proposed BP technique is as follows. First, PPG are divided into each beat. Then, 9 SCSA for beats. addition, 5 biometric data used. Biometrics include Heart Rate (HR), age, sex, height, weight. total 14 used training validating 1D-CNN model. After testing three types 1D-CNNs, model with most optimal performance selected. selected structure consists convolutional layers one fully connected layer. measured by Mean Error (ME) ± Standard Deviation (STD) following Association Advancement Medical Instrumentation (AAMI) standard. According to results test, Systolic (SBP) -2.99±14.48 mmHg Diastolic (DBP) 1.16±9.30 mmHg. Using technique, blood pressure can be easily predicted non-invasive cuff-less wearable sensor.