Open AccessDetection and Classification of ECG Signal through Machine learning
Author(s) -
Shalini Sahay,
A. K. Wadhwani,
Sulochana Wadhwani,
Sarita Singh Bhadauria
Publication year - 2019
Publication title -
international journal of innovative technology and exploring engineering
Language(s) - English
Resource type - Journals
ISSN - 2278-3075
DOI - 10.35940/ijitee.j1165.0881019
Subject(s) - particle swarm optimization , computer science , artificial neural network , pattern recognition (psychology) , artificial intelligence , signal (programming language) , sensitivity (control systems) , software , set (abstract data type) , anomaly detection , machine learning , engineering , electronic engineering , programming language
The electrical activity which might be acquired by inserting the probes on the body exterior that is originated within the individual muscle cells of the heart and is summed to indicate an indication wave form referred to as the EKG (ECG). Cardiac Arrhythmia is an associate anomaly within the heart which may be diagnosed with the usage of signals generated by Electrocardiogram (ECG). For the classification of ECG signals a software application model was developed and has been investigated with the usage of the MIT-BIH database. The version is based on some existing algorithms from literature, entails the extraction of a few temporal features of an ECG signal and simulating it with a trained FFNN. The software version may be employed for the detection of coronary heart illnesses in patients. The neural network’s structure and weights are optimized using Particle Swarm Optimization (PSO). The FFNN trained with set of rules by PSO increase its accuracy. The overall accuracy and sensitivity of the algorithm is about 93.687 % and 92%.