Fuzzy weighted-offset multi-scale edge detection for automatic echocardiographic lV boundary extraction

A. Sheikh Akbari, J.J. Soraghan

Research output: Contribution to conferencePaper

Abstract

This paper describes a new Fuzzy, Weighted-Offset, Multiscale Edge Detection algorithm for cardiac left ventricular (LV) epicardial and endocardial boundary detection on short axis (SA) echocardiographic images. The proposed method uses the `centre-based' approach, previously described in S.K. Setarehdan and J.J. Soraghan, 1999, IEEE Transaction on Biomedical Engineering. Vol. 46, No. 11, 1364 - 1378. The Edge-detection stage uses a new Fuzzy Weighted Offset Multiscale Edge Detection (FWOMED) technique in order to identify a single moving point for each one of the epicardial and endocardial boundaries over the N radii in an echocardiographic frame. This technique achieves optimal edge detection through non-decimated wavelet decomposition of the original signal followed by a fuzzy based decision technique, which is applied across the scales. Finally, a uniform cubic B-spline approximation is used to define the closed LV boundaries. The performance of this technique is compared to Mallat's (S. Mallat and S. Zhong, 1992, IEEE Trans. PAMI, Vol. 14, No. 7, 710-723.) multiscale edge detection technique, for a range of test data sets comprising different synthetic noisy signals.
LanguageEnglish
Pages1-6
Number of pages6
Publication statusPublished - 7 Oct 2002
EventIEE Seminar Medical Applications of Signal Processing - London, United Kingdom
Duration: 7 Oct 20027 Oct 2002

Conference

ConferenceIEE Seminar Medical Applications of Signal Processing
CountryUnited Kingdom
CityLondon
Period7/10/027/10/02

Fingerprint

Edge detection
Biomedical engineering
Wavelet decomposition
Splines

Keywords

  • fuzzy weighted
  • multi-scale
  • edge detection
  • automatic
  • echocardiographic lV boundary extraction

Cite this

Sheikh Akbari, A., & Soraghan, J. J. (2002). Fuzzy weighted-offset multi-scale edge detection for automatic echocardiographic lV boundary extraction. 1-6. Paper presented at IEE Seminar Medical Applications of Signal Processing , London, United Kingdom.
Sheikh Akbari, A. ; Soraghan, J.J. / Fuzzy weighted-offset multi-scale edge detection for automatic echocardiographic lV boundary extraction. Paper presented at IEE Seminar Medical Applications of Signal Processing , London, United Kingdom.6 p.
@conference{a1273595d6db4be790ba3a33f8b3b753,
title = "Fuzzy weighted-offset multi-scale edge detection for automatic echocardiographic lV boundary extraction",
abstract = "This paper describes a new Fuzzy, Weighted-Offset, Multiscale Edge Detection algorithm for cardiac left ventricular (LV) epicardial and endocardial boundary detection on short axis (SA) echocardiographic images. The proposed method uses the `centre-based' approach, previously described in S.K. Setarehdan and J.J. Soraghan, 1999, IEEE Transaction on Biomedical Engineering. Vol. 46, No. 11, 1364 - 1378. The Edge-detection stage uses a new Fuzzy Weighted Offset Multiscale Edge Detection (FWOMED) technique in order to identify a single moving point for each one of the epicardial and endocardial boundaries over the N radii in an echocardiographic frame. This technique achieves optimal edge detection through non-decimated wavelet decomposition of the original signal followed by a fuzzy based decision technique, which is applied across the scales. Finally, a uniform cubic B-spline approximation is used to define the closed LV boundaries. The performance of this technique is compared to Mallat's (S. Mallat and S. Zhong, 1992, IEEE Trans. PAMI, Vol. 14, No. 7, 710-723.) multiscale edge detection technique, for a range of test data sets comprising different synthetic noisy signals.",
keywords = "fuzzy weighted, multi-scale, edge detection, automatic , echocardiographic lV boundary extraction",
author = "{Sheikh Akbari}, A. and J.J. Soraghan",
year = "2002",
month = "10",
day = "7",
language = "English",
pages = "1--6",
note = "IEE Seminar Medical Applications of Signal Processing ; Conference date: 07-10-2002 Through 07-10-2002",

}

Sheikh Akbari, A & Soraghan, JJ 2002, 'Fuzzy weighted-offset multi-scale edge detection for automatic echocardiographic lV boundary extraction' Paper presented at IEE Seminar Medical Applications of Signal Processing , London, United Kingdom, 7/10/02 - 7/10/02, pp. 1-6.

Fuzzy weighted-offset multi-scale edge detection for automatic echocardiographic lV boundary extraction. / Sheikh Akbari, A.; Soraghan, J.J.

2002. 1-6 Paper presented at IEE Seminar Medical Applications of Signal Processing , London, United Kingdom.

Research output: Contribution to conferencePaper

TY - CONF

T1 - Fuzzy weighted-offset multi-scale edge detection for automatic echocardiographic lV boundary extraction

AU - Sheikh Akbari, A.

AU - Soraghan, J.J.

PY - 2002/10/7

Y1 - 2002/10/7

N2 - This paper describes a new Fuzzy, Weighted-Offset, Multiscale Edge Detection algorithm for cardiac left ventricular (LV) epicardial and endocardial boundary detection on short axis (SA) echocardiographic images. The proposed method uses the `centre-based' approach, previously described in S.K. Setarehdan and J.J. Soraghan, 1999, IEEE Transaction on Biomedical Engineering. Vol. 46, No. 11, 1364 - 1378. The Edge-detection stage uses a new Fuzzy Weighted Offset Multiscale Edge Detection (FWOMED) technique in order to identify a single moving point for each one of the epicardial and endocardial boundaries over the N radii in an echocardiographic frame. This technique achieves optimal edge detection through non-decimated wavelet decomposition of the original signal followed by a fuzzy based decision technique, which is applied across the scales. Finally, a uniform cubic B-spline approximation is used to define the closed LV boundaries. The performance of this technique is compared to Mallat's (S. Mallat and S. Zhong, 1992, IEEE Trans. PAMI, Vol. 14, No. 7, 710-723.) multiscale edge detection technique, for a range of test data sets comprising different synthetic noisy signals.

AB - This paper describes a new Fuzzy, Weighted-Offset, Multiscale Edge Detection algorithm for cardiac left ventricular (LV) epicardial and endocardial boundary detection on short axis (SA) echocardiographic images. The proposed method uses the `centre-based' approach, previously described in S.K. Setarehdan and J.J. Soraghan, 1999, IEEE Transaction on Biomedical Engineering. Vol. 46, No. 11, 1364 - 1378. The Edge-detection stage uses a new Fuzzy Weighted Offset Multiscale Edge Detection (FWOMED) technique in order to identify a single moving point for each one of the epicardial and endocardial boundaries over the N radii in an echocardiographic frame. This technique achieves optimal edge detection through non-decimated wavelet decomposition of the original signal followed by a fuzzy based decision technique, which is applied across the scales. Finally, a uniform cubic B-spline approximation is used to define the closed LV boundaries. The performance of this technique is compared to Mallat's (S. Mallat and S. Zhong, 1992, IEEE Trans. PAMI, Vol. 14, No. 7, 710-723.) multiscale edge detection technique, for a range of test data sets comprising different synthetic noisy signals.

KW - fuzzy weighted

KW - multi-scale

KW - edge detection

KW - automatic

KW - echocardiographic lV boundary extraction

M3 - Paper

SP - 1

EP - 6

ER -

Sheikh Akbari A, Soraghan JJ. Fuzzy weighted-offset multi-scale edge detection for automatic echocardiographic lV boundary extraction. 2002. Paper presented at IEE Seminar Medical Applications of Signal Processing , London, United Kingdom.