An approach of neural network based fetal ECG extraction

Md. Mamun Ibne Reaz, Lee Sze Wei

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper, we describe an adaptive method to separate fetal EGG from composite ECG that consists of both maternal and fetal ECGs by using ADALINE (Adaptive Linear Network). The input signal is the maternal signal and the target signal is the composite signal. The network emulate maternal signal as closely as possible to abdominal signal thus only predict the maternal ECG in the abdominal ECG. The network error equals abdominal ECG minus maternal ECG, which is the fetal ECG. The characteristic that enables fetal extraction is due to correlation between maternal ECG signals with the abdominal ECG signal of pregnant woman. A GUI program is written in Matlab to detect the changes in extracted fetal ECG by different values of momentum, learning rate and initial weights used in the network. However, the learning rate, momentum and initial weights are adjusted until the results are reasonably well. It is found that filtering performs best by high learning rate, low momentum, and small initial weights.

Original languageEnglish
Title of host publicationProceedings - 6th International Workshop on Enterprise Networking and Computing in Healthcare Industry, Healthcom 2004
EditorsK. Kurokawa, I. Nakajima, Y. Ishibashi
Pages57-60
Number of pages4
Publication statusPublished - 2004
EventProceedings - 6th International Workshop on Enterprise Networking and Computing in Healthcare Industry, Healthcom 2004 - Odawara
Duration: 28 Jun 200429 Jun 2004

Other

OtherProceedings - 6th International Workshop on Enterprise Networking and Computing in Healthcare Industry, Healthcom 2004
CityOdawara
Period28/6/0429/6/04

Fingerprint

Electrocardiography
Neural networks
Momentum
Linear networks
Composite materials
Graphical user interfaces

Keywords

  • Adaptive Linear Neural Network
  • ECG
  • Fetal ECG
  • Neural Network

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Ibne Reaz, M. M., & Wei, L. S. (2004). An approach of neural network based fetal ECG extraction. In K. Kurokawa, I. Nakajima, & Y. Ishibashi (Eds.), Proceedings - 6th International Workshop on Enterprise Networking and Computing in Healthcare Industry, Healthcom 2004 (pp. 57-60)

An approach of neural network based fetal ECG extraction. / Ibne Reaz, Md. Mamun; Wei, Lee Sze.

Proceedings - 6th International Workshop on Enterprise Networking and Computing in Healthcare Industry, Healthcom 2004. ed. / K. Kurokawa; I. Nakajima; Y. Ishibashi. 2004. p. 57-60.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Ibne Reaz, MM & Wei, LS 2004, An approach of neural network based fetal ECG extraction. in K Kurokawa, I Nakajima & Y Ishibashi (eds), Proceedings - 6th International Workshop on Enterprise Networking and Computing in Healthcare Industry, Healthcom 2004. pp. 57-60, Proceedings - 6th International Workshop on Enterprise Networking and Computing in Healthcare Industry, Healthcom 2004, Odawara, 28/6/04.
Ibne Reaz MM, Wei LS. An approach of neural network based fetal ECG extraction. In Kurokawa K, Nakajima I, Ishibashi Y, editors, Proceedings - 6th International Workshop on Enterprise Networking and Computing in Healthcare Industry, Healthcom 2004. 2004. p. 57-60
Ibne Reaz, Md. Mamun ; Wei, Lee Sze. / An approach of neural network based fetal ECG extraction. Proceedings - 6th International Workshop on Enterprise Networking and Computing in Healthcare Industry, Healthcom 2004. editor / K. Kurokawa ; I. Nakajima ; Y. Ishibashi. 2004. pp. 57-60
@inproceedings{245e2ab1357f4377b4eb1cad5ecd5224,
title = "An approach of neural network based fetal ECG extraction",
abstract = "In this paper, we describe an adaptive method to separate fetal EGG from composite ECG that consists of both maternal and fetal ECGs by using ADALINE (Adaptive Linear Network). The input signal is the maternal signal and the target signal is the composite signal. The network emulate maternal signal as closely as possible to abdominal signal thus only predict the maternal ECG in the abdominal ECG. The network error equals abdominal ECG minus maternal ECG, which is the fetal ECG. The characteristic that enables fetal extraction is due to correlation between maternal ECG signals with the abdominal ECG signal of pregnant woman. A GUI program is written in Matlab to detect the changes in extracted fetal ECG by different values of momentum, learning rate and initial weights used in the network. However, the learning rate, momentum and initial weights are adjusted until the results are reasonably well. It is found that filtering performs best by high learning rate, low momentum, and small initial weights.",
keywords = "Adaptive Linear Neural Network, ECG, Fetal ECG, Neural Network",
author = "{Ibne Reaz}, {Md. Mamun} and Wei, {Lee Sze}",
year = "2004",
language = "English",
isbn = "0780384539",
pages = "57--60",
editor = "K. Kurokawa and I. Nakajima and Y. Ishibashi",
booktitle = "Proceedings - 6th International Workshop on Enterprise Networking and Computing in Healthcare Industry, Healthcom 2004",

}

TY - GEN

T1 - An approach of neural network based fetal ECG extraction

AU - Ibne Reaz, Md. Mamun

AU - Wei, Lee Sze

PY - 2004

Y1 - 2004

N2 - In this paper, we describe an adaptive method to separate fetal EGG from composite ECG that consists of both maternal and fetal ECGs by using ADALINE (Adaptive Linear Network). The input signal is the maternal signal and the target signal is the composite signal. The network emulate maternal signal as closely as possible to abdominal signal thus only predict the maternal ECG in the abdominal ECG. The network error equals abdominal ECG minus maternal ECG, which is the fetal ECG. The characteristic that enables fetal extraction is due to correlation between maternal ECG signals with the abdominal ECG signal of pregnant woman. A GUI program is written in Matlab to detect the changes in extracted fetal ECG by different values of momentum, learning rate and initial weights used in the network. However, the learning rate, momentum and initial weights are adjusted until the results are reasonably well. It is found that filtering performs best by high learning rate, low momentum, and small initial weights.

AB - In this paper, we describe an adaptive method to separate fetal EGG from composite ECG that consists of both maternal and fetal ECGs by using ADALINE (Adaptive Linear Network). The input signal is the maternal signal and the target signal is the composite signal. The network emulate maternal signal as closely as possible to abdominal signal thus only predict the maternal ECG in the abdominal ECG. The network error equals abdominal ECG minus maternal ECG, which is the fetal ECG. The characteristic that enables fetal extraction is due to correlation between maternal ECG signals with the abdominal ECG signal of pregnant woman. A GUI program is written in Matlab to detect the changes in extracted fetal ECG by different values of momentum, learning rate and initial weights used in the network. However, the learning rate, momentum and initial weights are adjusted until the results are reasonably well. It is found that filtering performs best by high learning rate, low momentum, and small initial weights.

KW - Adaptive Linear Neural Network

KW - ECG

KW - Fetal ECG

KW - Neural Network

UR - http://www.scopus.com/inward/record.url?scp=14544301002&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=14544301002&partnerID=8YFLogxK

M3 - Conference contribution

SN - 0780384539

SP - 57

EP - 60

BT - Proceedings - 6th International Workshop on Enterprise Networking and Computing in Healthcare Industry, Healthcom 2004

A2 - Kurokawa, K.

A2 - Nakajima, I.

A2 - Ishibashi, Y.

ER -