Comparison of multiwavelet, wavelet, Haralick, and shape features for microcalcification classification in mammograms

Abstract

We present an evaluation and comparison of the performance of four di5erent texture and shape feature extraction methods for classi(cation of benign and malignant microcalci(cations in mammograms. For 103 regions containing microcalci(cation clusters, texture and shape features were extracted using four approaches: conventional shape quanti(ers; co-occurrence-based method of Haralick; wavelet transformations; and multi-wavelet transformations. For each set of features, most discriminating features and their optimal weights were found using real-valued and binary genetic algorithms (GA) utilizing a k-nearest-neighbor classi(er and a malignancy criterion for generating ROC curves for measuring the performance. The best set of features generated areas under the ROC curve ranging from 0.84 to 0.89 when using real-valued GA and from 0.83 to 0.88 when using binary GA. The multi-wavelet method outperformed the other three methods, and the conventional shape features were superior to the wavelet and Haralick features. ? 2004 Pattern Recognition Society. Published by Elsevier Ltd. All rights reserved.

DOI: 10.1016/j.patcog.2003.03.001

Extracted Key Phrases

21 Figures and Tables

0102030'05'06'07'08'09'10'11'12'13'14'15'16'17
Citations per Year

169 Citations

Semantic Scholar estimates that this publication has 169 citations based on the available data.

See our FAQ for additional information.

Cite this paper

@article{SoltanianZadeh2004ComparisonOM, title={Comparison of multiwavelet, wavelet, Haralick, and shape features for microcalcification classification in mammograms}, author={Hamid Soltanian-Zadeh and Farshid Rafiee-Rad and Siamak Pourabdollah-Nejad D}, journal={Pattern Recognition}, year={2004}, volume={37}, pages={1973-1986} }