Universiti Teknologi Malaysia Institutional Repository

Application of texture analysis in echocardiography images for myocardial infarction tissue

Agani, N. and Abu Bakar, Syed Abdul Rahman and Sheikh Salleh, Sheikh Hussain (2007) Application of texture analysis in echocardiography images for myocardial infarction tissue. Jurnal Teknologi (46D). pp. 61-76. ISSN 2180-3722

[img] PDF (Full Text) - Published Version
Restricted to Repository staff only

260kB
[img] HTML - Published Version
485kB

Abstract

Texture analysis is an important characteristic for surface and object identification from medical images and many other types of images. This research has developed an algorithm for texture analysis using medical images do trained from echocardiography in identifying heart with suspected myocardial infarction problem. A set of combination of wavelet extension transform with gray level co-occurrence matrix is proposed. In this work, wavelet extension transform is used to form an image approximation with higher resolution. The gray level co-occurrence matrices computed for each subband are used to extract four feature vectors: entropy, contrast, energy (angular second moment) and homogeneity (inverse difference moment). The classifier used in this work is the Mahalanobis distance classifier. The method is tested with clinical data from echocardiography images of 17 patients. For each patient, tissue samples are taken from suspected infarcted area as well as from non-infarcted (normal) area. For each patient, 8 frames separated by some time interval are used and for each frame, 5 normal regions and 5 suspected myocardial infarction regions of 16×16 pixel size are analyzed. The classification performance achieved 91.32% accuracy.

Item Type:Article
Subjects:T Technology > T Technology (General)
T Technology > TK Electrical engineering. Electronics Nuclear engineering
Divisions:Electrical Engineering
ID Code:8044
Deposited By: Norshiela Buyamin
Deposited On:25 Mar 2009 07:35
Last Modified:01 Nov 2017 04:17

Repository Staff Only: item control page