Publication detail

Comparison of Baseline Wandering Removal Algorithms for Automatic Clas sification of Electrocardiogram

SMÍŠEK, R. VÍTEK, M. KOLÁŘOVÁ, J.

Original Title

Comparison of Baseline Wandering Removal Algorithms for Automatic Clas sification of Electrocardiogram

Type

conference paper

Language

English

Original Abstract

Baseline wandering (sometimes also baseline drift) is a noise that exacerbates the evaluation of ECG and reduces the success of automatic ECG classifiers. In previous works, many methods have been proposed to remove such noise. The aim of this article is to compare commonly used methods. Each signal is in parallel reprocessed by all analyzed methods a nd then enters an automatic classifier that is able to classify QRS complexes. According to the classification success, the proposed methods were compared. Based on the results of this work, the best method for removing baseline wandering in the ECG is wav elet filtration. The success of the classification is further improved by the combination of wavelet filtration and EEMD. The disadvantage of this combination is its very high computational complexity .

Keywords

baseline wandering ; baseline drift ; ECG ; auto matic classification; EMD; EEMD; wavelet filtration

Authors

SMÍŠEK, R.; VÍTEK, M.; KOLÁŘOVÁ, J.

Released

28. 8. 2017

Publisher

Vysoké učení technické v Brně, Fakulta elektrotechniky a komunikačních technologií

Location

Mikulov

ISBN

978-80-214-5526-9

Book

Proceedings of IEEE Student Branch Conference Mikulov 2017

Edition number

1

Pages from

63

Pages to

66

Pages count

77

BibTex

@inproceedings{BUT139071,
  author="Radovan {Smíšek} and Martin {Vítek} and Jana {Kolářová}",
  title="Comparison of Baseline Wandering Removal Algorithms for Automatic Clas sification of Electrocardiogram",
  booktitle="Proceedings of IEEE Student Branch Conference Mikulov 2017",
  year="2017",
  number="1",
  pages="63--66",
  publisher="Vysoké učení technické v Brně, Fakulta elektrotechniky a komunikačních technologií",
  address="Mikulov",
  isbn="978-80-214-5526-9"
}