Data di Pubblicazione:
2014
Citazione:
A modified Bayesian filtering framework for ECG beat segmentation / A. Kheirati Roonizi, M. Fatemi - In: Electrical Engineering (ICEE), 2014 22nd Iranian Conference on[s.l] : IEEE, 2014. - ISBN 9781479944095. - pp. 1868-1872 (( Intervento presentato al 22. convegno Iranian Conference on Electrical Engineering tenutosi a Tehran nel 2014 [10.1109/IranianCEE.2014.6999844].
Abstract:
In this paper, we have presented a modified EKF structure based on the previously introduced signal decomposition based ECG Dynamic Model (EDM) for ECG beat segmentation. The new EKF can simultaneously estimate each of the ECG components including P, Q, R, S and T waveforms as well as the ECG signal. In this framework, instantaneous Gaussian functions of the P, Q, R, S and T components are considered as hidden state variables that are distinctly estimated from sample to sample. The result have shown that each of the CWs have been accurately estimated from multiple ECG beats. The proposed EDM can also be useful for synthetic ECG generation and ECG denoising applications.
Tipologia IRIS:
03 - Contributo in volume
Keywords:
ECG Dynamic Model; ECG delineation; Extended Kalman filtering; ECG Beat Segmentation
Elenco autori:
A. Kheirati Roonizi, M. Fatemi
Link alla scheda completa:
Titolo del libro:
Electrical Engineering (ICEE), 2014 22nd Iranian Conference on