Premature ventricular contraction detection using artificial neural network developed in android application

Abstract

We have conducted a study of detection system for premature ventricular contraction (PVC) developed in an android mobile phone. The system utilizes artificial neural network (ANN) with electrocardiographic (ECG) features of RR interval and QRS width. RR Interval and QRS width is Interval in ECG waveform. The algorithms of the detection are implemented using JAVA Eclipse Juno. The system is examined using electrocardiography of patients provided by Physionet MIT-BIH. The feature number is varied and the best result is found when both features RR interval and QRS width are applied with the performances of 94.58%, 96.59% and 96.29% in terms of sensitivity, specificity and accuracy Keywords-Premature ventricular contraction, artificial neural network and android device ;

Automatic Premature Ventricular Contraction Detection Using Deep Metric Learning and KNN

“…The essence of the classifier is a hypothesis or discrete-valued function. There are some popular classifiers used to distinguish regular and PVC beats: Artificial neural networks (ANN) [ 20 , 21 , 22 ], learning vector quantization neural network (LVQNN) [ 23 ], k-nearest neighbours (k-NN) algorithm [ 24 , 25 ], discrete hidden Markov model (DHMM) [ 26 ], support vector machine (SVM) [ 27 , 28 ], Bayesian classification algorithms [ 29 ], and random forest (RF) [ 30 ].…”

Detection of Premature Ventricular Contractions Using Densely Connected Deep Convolutional Neural Network with Spatial Pyramid Pooling Layer

“…Up until now, various methods for PVC detection have been developed with the aim of reducing doctors workload and these methods can be categorized into two classes.First,hand-craft features combining a classifier.These kind of methods are often seen in earlier published works that utilize hand-craft ECG features and a classifier to distinguish between PVC and non-PVC.The hand-craft features include morphology [9], wavelet [12,5,26,29,24] and temporal domain [18] etc. Some works also studied methods for the selection of handcraft features to improve detection performance [11,14,21].Various classifiers are used in the studies of PVC detection including artificial neural network [15,20,34],support vector machine [4] as well as clustering [3].Second, deep convolutional neural network (CNN)is showing advantages over traditional methods by providing a way of learning highly discriminative features automatically. Many works have studied its application in detecting abnormal heart beats including atrial fibrillation [2,27,30,7,25] and PVC [33,31,8] or other arrhythmia [22].…”