Wearable Wireless Sensors Network for ECG Telemonitoring Using Neural Network for Features Extraction

The technological progress of wireless communication, embedded systems and health offers innovative alternatives to medical care, in particular, telemonitoring and telediagnosis. ECG signal monitoring is a vital indicator in the control of heart disease. Nevertheless, one of the main challenges of remote monitoring of heart rate is the requirement of control in accordance with the service provided by hospital equipment. In this article, an approach to ECG telemonitoring based on wireless sensor networks combined with the Internet of Things (IoT) is proposed. The ECG signal is measured using a wearable sensor node allowing high-frequency noise suppression. The collected data is transmitted to the Gateway node, which performs complex processing including baseline and linear variations suppression using polynomial interpolation, extraction of R peaks using the Multilayer Perceptron Neural Network. It can determine the variation in heart rate by the using the extracted R signal. Thanks to IoT technology, the Gateway node is able to aggregate data into an IoT platform  through an IoT cloud for visual telemonitoring of heart rate in real-time. The experimental results show that the system is effective and reliable for the collection, transmission, and display of ECG data in real time for the purpose of telemonitoring of patients with heart disease.