Atrial Fibrillation ECG Signal Processing for QRST Cancellation: Zero-Padding versus Time Alignment

S Gholinezhadasnefestani, OJ Escalona, NC Castro, K Nazarzadeh, V Kodoth, E Lau, G Manoharan

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In non-invasive analysis of atrial fibrillation (AF), the QRST cancellation is commonly applied by using a time-aligned average over several ECG cycles. This paper proposes another technique with less illusive high frequency content into the residual signal. Firstly the QRS complexes and T-end were detected by Pan-Tompkins's technique and Zhang's algorithm respectively. Subsequently, the QRST segment was determined in eachperiod of the ECG signal. Then, two different approaches (time-aligned and zero padding) were applied to remove the QRST segments from each period and make the signalready for AF analysis. To compare the effectiveness of the QRST cancellation approaches, mean spectral power of the residual signal at high frequency bands (upper 12 Hz)was used. The results indicated a superiority of the timealigned averaging method for the application of AF analysis.
Original languageEnglish
Title of host publicationUnknown Host Publication
PublisherIEEE Computer Society
Pages45-48
Number of pages4
ISBN (Print)978-0-7695-4882-1/12
DOIs
Publication statusPublished - 15 Dec 2012
EventIEEE Andean Region International Conference 2012 (Andescon 2012) - Cuenca, Ecuador
Duration: 15 Dec 2012 → …
http://andescon2012.org/index.php/congreso2012/andesconcuenca

Conference

ConferenceIEEE Andean Region International Conference 2012 (Andescon 2012)
Period15/12/12 → …
Internet address

Keywords

  • QRS cancellation
  • ECG
  • cardiology
  • frequency analysis
  • signal averaging
  • atrial fibrillation
  • AF cardioversion
  • zero padding
  • power density spectrum.

Fingerprint Dive into the research topics of 'Atrial Fibrillation ECG Signal Processing for QRST Cancellation: Zero-Padding versus Time Alignment'. Together they form a unique fingerprint.

Cite this