ResearchBib Share Your Research, Maximize Your Social Impacts
Sign for Notice Everyday Sign up >> Login

MOBILE APP BASED REAL TIME ECG REMOTE MONITORING SYSTEM USING MACHINE LEARNING TECHNIQUE

Journal: International Journal of Advanced Research in Engineering and Technology (IJARET) (Vol.12, No. 01)

Publication Date:

Authors : ;

Page : 209-229

Keywords : Electrocardiogram; Arrhythmia; Artificial Neural Network; Accuracy; MATLAB; Wireless network and MongoDB.;

Source : Downloadexternal Find it from : Google Scholarexternal

Abstract

Background and objective: The term arrhythmia describe by irregularity of heart beat. Addiction of alcohol, hypertension, and mental pressure are main cause behind it. Palpitations, dizziness, and chest pain are the symptoms of a patients who suffer from arrhythmia. Physicians are diagnosis this illness by regular checking patient medical history, diet and lifestyle. Methods: We propose an automated classification scheme which is an applied part of Artificial Neural Network (ANN) to detect the cardiac arrhythmia, using parallel ECG recordings. We mainly concentrate on some major points hence de-nosing unprocessed ECG data, then generate an authenticate classification for arrhythmia and develop an app based monitoring system associated with MongoDB server. Various detection points of ECG signal namely P, Q, R and S waveforms has under consideration during evaluation and classification using neural network model with Back Propagation algorithm. This classification has done between two classes named normal and abnormal. To get a better result here we consider two different datasets for training and testing purpose network model, one is real data set named as MYECGDB with 600 sample and Physionet database. Results: Resultant data is provided in this paper that, for the detection of Arrhythmia, the Back propagation algorithm with ANN gives 96.50% accuracy on MYECGDB and 95.60% on MIT-BIH database along with NSR database. Conclusions: The proposed method is able to provide perfect diagnosis of the heart in wireless environment and the method is based on the feature selection, classification of the raw ECG signals.

Last modified: 2021-03-25 17:00:09