Detail publikace

Detection of QRS Notching Using Continuous Wavelet Transform

Originální název

Detection of QRS Notching Using Continuous Wavelet Transform

Anglický název

Detection of QRS Notching Using Continuous Wavelet Transform

Jazyk

en

Originální abstrakt

QRS notching is an indicator of myocardial ischemia, early repolarization and according to the new criteria also left bundle branch block (LBBB). Automatic detection of QRS notching is therefore highly desirable. The aim of this article is the automatic detection of QRS notching using a continuous wavelet transform (CWT). The Haar wavelet was used as the mother wavelet and the scales were chosen in the range from 6 to 70. For testing were used 123 signals from the CSE database (12 leads, 10 seconds duration, 500 Hz sampling frequency and 16 bits resolution). The automatic detection had the sensitivity of 88.07%, the specificity of 94.23% and the positive predictive value of 93.16%. The accuracy is 91.33%.

Anglický abstrakt

QRS notching is an indicator of myocardial ischemia, early repolarization and according to the new criteria also left bundle branch block (LBBB). Automatic detection of QRS notching is therefore highly desirable. The aim of this article is the automatic detection of QRS notching using a continuous wavelet transform (CWT). The Haar wavelet was used as the mother wavelet and the scales were chosen in the range from 6 to 70. For testing were used 123 signals from the CSE database (12 leads, 10 seconds duration, 500 Hz sampling frequency and 16 bits resolution). The automatic detection had the sensitivity of 88.07%, the specificity of 94.23% and the positive predictive value of 93.16%. The accuracy is 91.33%.

BibTex


@inproceedings{BUT127527,
  author="Radovan {Smíšek} and Martin {Vítek}",
  title="Detection of QRS Notching Using Continuous Wavelet Transform",
  annote="QRS notching is an indicator of myocardial ischemia, early repolarization and according to the new criteria also left bundle branch block (LBBB). Automatic detection of QRS notching is therefore highly desirable. The aim of this article is the automatic detection of QRS notching using a continuous wavelet transform (CWT). The Haar wavelet was used as the mother wavelet and the scales were chosen in the range from 6 to 70. For testing were used 123 signals from the CSE database (12 leads, 10 seconds duration, 500 Hz sampling frequency and 16 bits resolution). The automatic detection had the sensitivity of 88.07%, the specificity of 94.23% and the positive predictive value of 93.16%. The accuracy is 91.33%.",
  address="Vysoké učení technické v Brně, Fakulta elektrotechniky a komunikačních technologií",
  booktitle="Sborník příspěvků studentské konference Blansko 2016",
  chapter="127527",
  howpublished="online",
  institution="Vysoké učení technické v Brně, Fakulta elektrotechniky a komunikačních technologií",
  year="2016",
  month="august",
  pages="83--86",
  publisher="Vysoké učení technické v Brně, Fakulta elektrotechniky a komunikačních technologií",
  type="conference paper"
}