A wireless sensor network for remote detection of arrhythmias using convolutional neural network

Wireless Sensor Network (WSN) is getting a lot of interest from governments, citizens, industries, and universities for applications in various domains like smart buildings, environmental surveillance, and healthcare. Designers of the WSNs must take care of certain prevalent issues associated with security, detection of faults, scheduling of events, energy-aware routing, clustering of nodes, and aggregation of data. In this work, a WSN based healthcare application for detecting Arrhythmia remotely is presented. The Electrocardiogram (ECG) is employed as the principal diagnostic tool for arrhythmia. The ECG signals are composed of information related to the distinct arrhythmia types. Nevertheless, these signals' non-linearity, as well as complexity, makes their manual analysis quite challenging. The majority of the researchers have a lot of interest in energy efficiency as almost all advancements in diverse technologies will eventually result in a sustainable global energy system. This work has employed the Artificial Bee Colony as well as the Grey Wolf Optimiser for optimizing the clustering to boost the routing and also the network longevity. This work presents a Convolutional Neural Network technique for automatically detecting the distinct ECG segments.