Design an Intelligent Real Time ECG Monitoring System Using Convolution Neural Network

Abstract

Electrocardiogram (ECG) monitoring is now becoming part of everyday health life. Through ECG characteristics such as patient's heartbeats, heart conditions, and heart disease can be analyzed. This paper presents the design and implementation of a system for analyzing and filtering the ECG signal and allowing its remote monitoring based on the use of deep learning algorithms, this algorithm is Convolution Neural Network (CNN), where the network was built in MATLAB and training using the dataset (PhysioNet 2017). when, the ESP NODE MCU microcontroller was used with the AD8232 sensor in designing a system that records the ECG signal from the patient in real time and filtering it using FIR filter that will be designed in MATLAB, then transmits it to the network that has been trained to be classified as whether it is normal or abnormal. Then, this result is transmitted locally to be displayed in monitoring side, the results showed high accuracy in classifying the signal and in filtering different Noise, as well as its speed in responding to a change in the condition of the signal and giving a warning to the observer. This contributes to speeding up the detection of the deterioration of the patient's condition in a timely manner.