Presentation 1997/9/19
Nonlinear Prediction Coding of ECG
Masaomi KOIZUMI, Noritaka MATUOKA, Yasunari YOKOTA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) A coding method for the compression of discrete electrocardiogram (ECG) is proposed. The method is based on nonlinear prediction in which Volterra functional series model or 3 layered Neural Network Model is employed as a nonlinear autoregressive model. It has been shown that there is no substantial improvement of prediction error even if using greater than 2 or 3 prediction order in linear prediction of ECG. Our numerical experiments indicate that the proposed nonlinear prediction technique extremely decrease the prediction error in QRS complex. It means that QRS complex includes nonlinearities, which could not be sufficiently eliminated by conventional linear prediction. The designed ECG coder can provide less bit rate, or higher performances than both linear prediction and wavelet orthogonal transform coders.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Volterra functional series / Neural network / Nonlinear prediction / Prediction coding / Data compression
Paper # NLP97-79
Date of Issue

Conference Information
Committee NLP
Conference Date 1997/9/19(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Nonlinear Problems (NLP)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Nonlinear Prediction Coding of ECG
Sub Title (in English)
Keyword(1) Volterra functional series
Keyword(2) Neural network
Keyword(3) Nonlinear prediction
Keyword(4) Prediction coding
Keyword(5) Data compression
1st Author's Name Masaomi KOIZUMI
1st Author's Affiliation Electronics and Computer Engineering Division, Graduate School of Engineering, Gifu University()
2nd Author's Name Noritaka MATUOKA
2nd Author's Affiliation Electronics and Computer Engineering Division, Graduate School of Engineering, Gifu University
3rd Author's Name Yasunari YOKOTA
3rd Author's Affiliation Department of Information Science, Faculty of Engineering, Gifu University
Date 1997/9/19
Paper # NLP97-79
Volume (vol) vol.97
Number (no) 258
Page pp.pp.-
#Pages 8
Date of Issue