ANN compression of morphologically similar ECG complexes

D J Hamilton, D C. Thomson, W.A Sandham

Research output: Contribution to journalArticle

Abstract

A compression algorithm for electrocardiogram signals is presented, based on an auto-associative neural network. Issues of weight and activation coding are considered, and compression performances of various network sizes are compared. A unique feature is the performance improvement achieved using DC level removal. A comparison with existing techniques is provided.
Original languageEnglish
Pages (from-to)841-843
Number of pages3
JournalMedical and Biological Engineering and Computing
Volume33
Issue number6
Publication statusPublished - Nov 1995
Externally publishedYes

Fingerprint

Electrocardiography
Chemical activation
Neural networks

Keywords

  • compression
  • electrocardiogram
  • neural nets

Cite this

@article{b25827eb3991496bb08e07272b95ec3c,
title = "ANN compression of morphologically similar ECG complexes",
abstract = "A compression algorithm for electrocardiogram signals is presented, based on an auto-associative neural network. Issues of weight and activation coding are considered, and compression performances of various network sizes are compared. A unique feature is the performance improvement achieved using DC level removal. A comparison with existing techniques is provided.",
keywords = "compression, electrocardiogram, neural nets",
author = "Hamilton, {D J} and Thomson, {D C.} and W.A Sandham",
year = "1995",
month = "11",
language = "English",
volume = "33",
pages = "841--843",
journal = "Medical and Biological Engineering and Computing",
issn = "1040-0118",
publisher = "Springer",
number = "6",

}

ANN compression of morphologically similar ECG complexes. / Hamilton, D J; Thomson, D C.; Sandham, W.A .

In: Medical and Biological Engineering and Computing, Vol. 33, No. 6, 11.1995, p. 841-843.

Research output: Contribution to journalArticle

TY - JOUR

T1 - ANN compression of morphologically similar ECG complexes

AU - Hamilton, D J

AU - Thomson, D C.

AU - Sandham, W.A

PY - 1995/11

Y1 - 1995/11

N2 - A compression algorithm for electrocardiogram signals is presented, based on an auto-associative neural network. Issues of weight and activation coding are considered, and compression performances of various network sizes are compared. A unique feature is the performance improvement achieved using DC level removal. A comparison with existing techniques is provided.

AB - A compression algorithm for electrocardiogram signals is presented, based on an auto-associative neural network. Issues of weight and activation coding are considered, and compression performances of various network sizes are compared. A unique feature is the performance improvement achieved using DC level removal. A comparison with existing techniques is provided.

KW - compression

KW - electrocardiogram

KW - neural nets

M3 - Article

VL - 33

SP - 841

EP - 843

JO - Medical and Biological Engineering and Computing

JF - Medical and Biological Engineering and Computing

SN - 1040-0118

IS - 6

ER -