Biocapteurs et bioélectronique

A Real-Time framework to monitor the Heart Rate based on PhotoPlethysmoGraphy (PPG) Sensors at all possible conditions

Abstract

P Madhan Mohan*, S Naresh, G Viswanathan1, CH Sai Mahesh

Heart Rate (HR) is one of the vital parameters in health care monitoring. Many methods are being proposed to estimate the HR in different cases, which are not compatible with real-time applications. The main objective of this work is to develop a framework for determining HR under all possible conditions using the PPG signal. The proposed work is the combination of time domain, frequency domain, and tracking using accelerometer data. This real-time framework is being developed and tested in real-time with cortex M4 board using the Keil μ vision 4 IDE. This framework detects HR under rest and motion conditions with minimum hardware resources and the accuracy of the HR in real time environment is 98.6 %. The Mean Absolute Percentage Error (MAPE) is 1.67 %, the Mean Absolute Error (MAE) is 1.18 BPM and the Reference Closeness Factor (RCF) is 0.987. The hardware implementation of the algorithm requires total of 47KB memory which runs at a speed of 5.7 MIPS.