Monitoring of HR and BP using DFT and Pan-Tompkins Algorithm

Abstract

In this paper we propose monitoring of blood pressure and heart rate using Discrete Fourier Transform and Pan-Tompkins algorithm to achieve higher wear ability and high accuracy. Motion artifacts induced by the head movements are deals with machine learning framework to enable practical application scenarios. Here we suggest to place all the electrocardiogram (ECG) and photoplethysmography (PPG) sensors behind two ears to successfully acquire weak ear -ECG/PPG signals using a semi customized platform. After introducing head motions towards, we apply an unsupervised learning algorithm, Pan Tompkins to learn and identify raw heartbeats from motion artifacts compacted signals. Furthermore, we propose another unsupervised learning algorithm to filter out distorted/faking heartbeats, for the estimation of ECG to PPG pulse transit time (PTT) and HR. Specifically, we introduce a Discrete Fourier Transform (DFT) to quantify distortion conditions of raw heartbeats referring to a high-quality heartbeat pattern, which are then compared with a threshold to perform purification. The heartbeat pattern and the distortion threshold are learned by a K-medoids clustering approach and a histogram triangle method, respectively. Afterwards, we perform a comparative analysis on ten PTT or PTT&HR-based BP learning models.