Telemonitoring Device of Blood Pressure and Heart Rate Through Multilayer Perceptrons and Pulse Rate Variability

Abstract

This paper presents the development of an electronic prototype for estimating systolic blood pressure (SBP), diastolic blood pressure (DBP) and heart rate (HR) by using the features extracted from the oscillometric signal obtained by the device. Firstly, HR was determined through the peaks provided by the oscillometric signal in a period of thirty seconds, providing an estimated value of beats per minute (bpm). Second, the blood pressure (BP) indexes were acquired using a multi-layer perceptron (MLP) and the combination of the Maximum Amplitude Algorithm (MAA), which detects the Mean Arterial Pressure (MAP). The estimated heart rate and the MAP were added to the MLP input to determine the diastolic and sistolic blood pressure. The parameters obtained were transmitted by a LPWAN module to a cloud database, visualizing the information via web server. The parameters obtained during the test phase were compared against Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE) provided by the literature, identifying a better response of DBP determination. The proposed prototype could be an auxiliary tool for the estimation of blood pressure and heart rate in a period less than 60 seconds for patients in ambulatory conditions helping the medical staff to obtain clinical information of patients during their daily activities.