Advanced

Wavelet transform as signal processing method for ANN-classification of acute coronary syndrome

Åström, Hampus LU (2011) FYTK01 20111
Computational Biology and Biological Physics
Abstract
When diagnosing acute coronary syndrome, time is of the essence and electrocardiography the most effective method to obtain data about the patients condition. Artificial neural networks can here be used to assist medical doctors.
In this text the wavelet transform is introduced as a signal processing method for piping ECG-curves to ANN’s. The performance of this setup (ANN performance is measured by AUC, the area under the receiver operating characteristic curve) varies around 0.60-0.65, significantly worse than that of previous methods, with an AUC of over 0.80. Some insight into ECG-curve properties and the compressional values of wavelet transforms is gained however.
Please use this url to cite or link to this publication:
author
Åström, Hampus LU
supervisor
organization
course
FYTK01 20111
year
type
M2 - Bachelor Degree
subject
keywords
wavelet transform, artificial neural networks, acute coronary syndrome, medical decision support
language
English
id
2204713
date added to LUP
2011-11-21 08:49:10
date last changed
2017-10-06 16:47:25
@misc{2204713,
  abstract     = {When diagnosing acute coronary syndrome, time is of the essence and electrocardiography the most effective method to obtain data about the patients condition. Artificial neural networks can here be used to assist medical doctors. 
In this text the wavelet transform is introduced as a signal processing method for piping ECG-curves to ANN’s. The performance of this setup (ANN performance is measured by AUC, the area under the receiver operating characteristic curve) varies around 0.60-0.65, significantly worse than that of previous methods, with an AUC of over 0.80. Some insight into ECG-curve properties and the compressional values of wavelet transforms is gained however.},
  author       = {Åström, Hampus},
  keyword      = {wavelet transform,artificial neural networks,acute coronary syndrome,medical decision support},
  language     = {eng},
  note         = {Student Paper},
  title        = {Wavelet transform as signal processing method for ANN-classification of acute coronary syndrome},
  year         = {2011},
}