SVM approach to classifying lesions in USG images with the use of the gabor decomposition

Authors:
Marcin Ciecholewski
Affiliations:
Institute of Computer Science, Jagiellonian University, Kraków, Poland
Venue:
IDEAL'11 Proceedings of the 12th international conference on Intelligent data engineering and automated learning
Year:
2011

Citing 8
Cited 0

The nature of statistical learning theory

The nature of statistical learning theory
Fast training of support vector machines using sequential minimal optimization

Advances in kernel methods
A Tutorial on Support Vector Machines for Pattern Recognition

Data Mining and Knowledge Discovery
SVM feature selection for classification of SPECT images of Alzheimer's disease using spatial information

Knowledge and Information Systems - Special Issue on Mining Low-Quality Data
Gallbladder boundary segmentation from ultrasound images using active contour model

IDEAL'10 Proceedings of the 11th international conference on Intelligent data engineering and automated learning
Gallbladder segmentation in 2-D ultrasound images using deformable contour methods

MDAI'10 Proceedings of the 7th international conference on Modeling decisions for artificial intelligence
Support vector machine approach to cardiac SPECT diagnosis

IWCIA'11 Proceedings of the 14th international conference on Combinatorial image analysis
Selecting regions of interest in SPECT images using wilcoxon test for the diagnosis of alzheimer's disease

HAIS'10 Proceedings of the 5th international conference on Hybrid Artificial Intelligence Systems - Volume Part I

Quantified Score

Hi-index	0.01

Visualization

Abstract

The article presents the application of the support vector machines (SVM) method to recognise gallbladder lesions such as lithiasis and polyps in USG images. USG images of the gallbladder were first processed by the histogram normalisation transformation to improve their contrast, and the gallbladder shape was segmented using active contour models. Then the background area of uneven contrast was eliminated from images. To extract features from the images to be classified, the Gabor decomposition was applied to a plane presented in a log-polar system. In the best case, the SVM achieved the accuracy of 82% for all lesions, 85.7% for lithiasis and 74.3% for polyps.