ECG Beat Classification Using Waveform Similarity and RR Intervals

作者：Ka Ahmad Khoureich^*

来源：Journal of Medical and Biological Engineering, 2012, 32(6): 417-421.

DOI：10.5405/jmbe.905

• electrocardiogram (ECG)
• beat classification
• RR interval
• Waveform transform
• Patient adaptation

This paper presents an electrocardiogram (ECG) beat classification method based on waveform similarity and RR intervals. The method classifies six types of heart beat, namely normal beats, atrial premature beats, paced beats, premature ventricular contractions, left bundle branch block beats, and right bundle branch block beats. The ECG signal is first denoised using wavelet-transform-based techniques. Heart beats, sampled at 128 points centered on the R peak, are extracted from the ECG signal. The number of samples per heart beat is then reduced to 16 to constitute a feature. The RR intervals surrounding the beat are also used as a feature. A database of annotated beats is built for the classifier for waveform comparison with unknown beats. The method achieved a classification rate of 97.52% on tests with 46 recordings of the MIT/BIH arrhythmia database.

• 出版日期2012