Denoising and Analysis of ECG Signal using Wavelet Transform for Detection of Arrhythmia
Shilpa Hudnurkar1, Ankita Wanchoo2
1Shilpa Hudnurkar*, Department of Electronics and Telecommunication, Symbiosis Institute of Technology, Symbiosis International (Deemed University), Pune, India.
2Ankita Wanchoo, Department of Electronics and Telecommunication, Symbiosis Institute of Technology, Symbiosis International (Deemed University), Pune, India.
Manuscript received on November 15, 2019. | Revised Manuscript received on November 23, 2019. | Manuscript published on November 30, 2019. | PP: 2113-2117 | Volume-8 Issue-4, November 2019. | Retrieval Number: D7683118419/2019©BEIESP | DOI: 10.35940/ijrte.D7683.118419
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Electrocardiography is fundamental in the observation of heart function and diagnosis of diseases related to it. It involves measurement of very small bioelectric signals (in millivolts) produced by the human heart during its opening and closing of valves in atria and ventricle and is represented on a scaled paper. P, QRS, and T wave annotations by cardiologists then help in the diagnosis of the patient. Due to the electrical activity of muscles (EMG), instability of electrode-skin contact and patient movement, the noise gets induced during the plotting of the electrocardiogram (ECG). It is important to remove the noise from this signal as it is a signal having very small amplitude and different frequencies repeated almost every second. For such nonstationary biosignals, Wavelet Transform (WT) can be used. In this study, Continuous Wavelet Transform (CWT) and Discrete Wavelet Transform (DWT) are used to denoise and extract features from the ECG, respectively. The features extracted from DWT are used as input to Artificial Neural Network (ANN) for the classification of normal and abnormal ECG. Abnormal ECGs are further classified into tachycardia and bradycardia. The results show that ANN can classify ECGs with high accuracy. The data used for this study is from the MIT-BIH Arrhythmia Database Directory.
Keywords: Arrhythmia, ANN, CWT, Denoising, DWT.
Scope of the Article: Aggregation, Integration, and Transformation.