Open AccessEnsemble empirical mode decomposition‐based optimised power line interference removal algorithm for electrocardiogram signal
Author(s) -
Jebaraj Jenitta,
Arumugam Rajeswari
Publication year - 2016
Publication title -
iet signal processing
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.384
H-Index - 42
eISSN - 1751-9683
pISSN - 1751-9675
DOI - 10.1049/iet-spr.2015.0292
Subject(s) - hilbert–huang transform , computer science , algorithm , pattern recognition (psychology) , preprocessor , artificial intelligence , interference (communication) , noise (video) , speech recognition , signal (programming language) , filter (signal processing) , computer vision , channel (broadcasting) , programming language , computer network , image (mathematics)
This study proposes an optimised algorithm to remove power line interference (PLI) from electrocardiogram (ECG) signal based on ensemble empirical mode decomposition (EEMD). A computationally efficient algorithm is one of the important requirements for real‐time monitoring of cardio activities and diagnosis of arrhythmias. Computational complexity in EEMD is significantly reduced by using the EMD as the preprocessing stage. The noisy ECG signal is decomposed into intrinsic mode functions (IMFs) using EMD. ECG signals which are affected by PLI are automatically identified based on the simple ratio of the zero crossing number of IMF components. EEMD is used to decompose only ECG segments constructed from the noisy IMF components. The proposed algorithm is evaluated by real ECG signals available in MIT‐BIH arrhythmia database in terms of signal‐to‐noise ratio and root mean square error. The computational efficiency of this new framework is measured using MATLAB profiling functions and compared with EMD, EEMD, sign‐based adaptive and EMD with wavelet‐based methods. Results show that the proposed algorithm performs better than the EMD, EEMD, sign‐based adaptive and EMD with wavelet‐based methods and it is computationally more efficient than EMD and EEMD methods.