Model-based fiducial points extraction for baseline wandered electrocardiograms

Sayadi, O ; Sharif University of Technology | 2008

459 Viewed
  1. Type of Document: Article
  2. DOI: 10.1109/TBME.2007.899302
  3. Publisher: 2008
  4. Abstract:
  5. A fast algorithm based on the nonlinear dynamical model for the electrocardiogram (ECG) is presented for the precise extraction of the characteristic points of these signals with baseline drift. Using the adaptive bionic wavelet transform, the baseline wander is removed efficiently. In fact by the means of the bionic wavelet transform, the resolution in the time-frequency domain can be adaptively adjusted not only by the signal frequency but also by the signal instantaneous amplitude and its first-order differential, which results in a better baseline wander cancellation. At the next step the parameters of the model are chosen to have the least square error with the original ECG. Determining the precise position of the waveforms of an ECG signal with baseline wander is complicated due to the varying amplitudes of its waveforms, the ambiguous and changing form of the complex and the unknown drift. A model-based approach handles these complications, therefore a method based on this concept has been developed and the fiducial points are accurately detected using the center and spread parameters of Gaussian-functions of the model. Simulation results show that the proposed method has an average sensitivity of 99.58%, average detection accuracy of 99.64%, and specificity of 100%. © 2006 IEEE
  6. Keywords:
  7. Algorithms ; Frequency domain analysis ; Nonlinear systems ; Physiological models ; Time domain analysis ; Wavelet transforms ; Baseline wander ; Bionic wavelet transf ; ECG dynamical models ; Fiducial points ; Accuracy ; Analytical error ; Analytical parameters ; Dynamics ; Electrocardiogram ; Frequency analysis ; Model ; Normal distribution ; Regression analysis ; Sensitivity and specificity ; Signal detection ; Signal transduction ; Waveform ; Algorithms ; Arrhythmias, Cardiac ; Artifacts ; Artificial Intelligence ; Computer Simulation ; Diagnosis, Computer-Assisted ; Electrocardiography ; Heart Rate ; Humans ; Models, Cardiovascular ; Pattern Recognition, Automated
  8. Source: IEEE Transactions on Biomedical Engineering ; Volume 55, Issue 1 , 2008 , Pages 347-351 ; 00189294 (ISSN)
  9. URL: https://ieeexplore.ieee.org/document/4360041