One of the important analyses of electrical activity of the heart is through electrocardiogram (ECG). The major and important step of this analysis is the extraction of pure ECG signals from noise which is an ongoing challenge. During recent years, many algorithms have been used for denoising the ECG signals. One of the well-known algorithms is nonlocal means (NLMs). The NLM is based on self-similarity of the patches. It is also based on the periodicity of the signal. Since ECG signal repeats its characteristics for regular intervals of time, NLM can be used for denoising ECG signals. This paper presents the comprehensive analysis of NLM algorithm on a wide set of well-known ECG data. The various ECG signals from ECG 101 to ECG 109 have been examined at different dB of the noise level. The analysis of NLM algorithm is carried out using various parameters such as signal-to-noise ratio (SNR), mean square error (MSE), and percent distortion (PRD). The paper further presents the numerical and graphical results to show the effectiveness of NLM algorithm.
CITATION STYLE
Kulkarni, S., & Ali, L. (2019). Comprehensive Analysis of Nonlocal Means Algorithm on Wide Set of ECG Signals. In Lecture Notes in Networks and Systems (Vol. 40, pp. 571–580). Springer. https://doi.org/10.1007/978-981-13-0586-3_56
Mendeley helps you to discover research relevant for your work.