Mass detection in mammography using principle component analysis and stepwise selection

Ping Sung Liao, Shu-Mei Guo, Nan Sue Yu, Cheng Yi Chen, San Kan Lee, Chein I. Chang

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

This paper aims at constructing a feasible computer-aided mass detection system in mammography. To effectively distinguish mass tissues from normal tissues, the tissue characteristics are represented by a set of features, including fractal dimension, compactness, gray level histogram moments, statistics of spatial gray level dependence, texture spectrum and texture feature coding method, respectively. With reference to the energy viewpoint of principle component analysis (PCA), we explore possible linear combinations of the raw 212 feature variables, or from the original feature space we will extract several feature variables whose data distribution is complete enough to model that of the original feature space. Experimental results show that PCA is a useful tool to restrict the size of the independent feature variables, and stepwise selection approach in conjunction with PCA can sieve effective feature variables out from the raw feature space. In particular, on applying probabilistic neural network on the filtered feature vectors it has achieved reasonably good performance. The correct classification rates of fatty tissue, fatty-glandular and dense-glandular are 98.4%, 89.6% and 87.8%, respectively.

Original languageEnglish
Pages (from-to)275-287
Number of pages13
JournalChinese Journal of Radiology
Volume31
Issue number6
Publication statusPublished - 2006 Dec 1

Fingerprint

Mammography
Fractals
Adipose Tissue

All Science Journal Classification (ASJC) codes

  • Radiology Nuclear Medicine and imaging

Cite this

Liao, P. S., Guo, S-M., Yu, N. S., Chen, C. Y., Lee, S. K., & Chang, C. I. (2006). Mass detection in mammography using principle component analysis and stepwise selection. Chinese Journal of Radiology, 31(6), 275-287.
Liao, Ping Sung ; Guo, Shu-Mei ; Yu, Nan Sue ; Chen, Cheng Yi ; Lee, San Kan ; Chang, Chein I. / Mass detection in mammography using principle component analysis and stepwise selection. In: Chinese Journal of Radiology. 2006 ; Vol. 31, No. 6. pp. 275-287.
@article{186316f2533145258a62ccd8e1e842b6,
title = "Mass detection in mammography using principle component analysis and stepwise selection",
abstract = "This paper aims at constructing a feasible computer-aided mass detection system in mammography. To effectively distinguish mass tissues from normal tissues, the tissue characteristics are represented by a set of features, including fractal dimension, compactness, gray level histogram moments, statistics of spatial gray level dependence, texture spectrum and texture feature coding method, respectively. With reference to the energy viewpoint of principle component analysis (PCA), we explore possible linear combinations of the raw 212 feature variables, or from the original feature space we will extract several feature variables whose data distribution is complete enough to model that of the original feature space. Experimental results show that PCA is a useful tool to restrict the size of the independent feature variables, and stepwise selection approach in conjunction with PCA can sieve effective feature variables out from the raw feature space. In particular, on applying probabilistic neural network on the filtered feature vectors it has achieved reasonably good performance. The correct classification rates of fatty tissue, fatty-glandular and dense-glandular are 98.4{\%}, 89.6{\%} and 87.8{\%}, respectively.",
author = "Liao, {Ping Sung} and Shu-Mei Guo and Yu, {Nan Sue} and Chen, {Cheng Yi} and Lee, {San Kan} and Chang, {Chein I.}",
year = "2006",
month = "12",
day = "1",
language = "English",
volume = "31",
pages = "275--287",
journal = "Chinese Journal of Radiology",
issn = "1018-8940",
publisher = "Radiological Society of the R.O.C.",
number = "6",

}

Liao, PS, Guo, S-M, Yu, NS, Chen, CY, Lee, SK & Chang, CI 2006, 'Mass detection in mammography using principle component analysis and stepwise selection', Chinese Journal of Radiology, vol. 31, no. 6, pp. 275-287.

Mass detection in mammography using principle component analysis and stepwise selection. / Liao, Ping Sung; Guo, Shu-Mei; Yu, Nan Sue; Chen, Cheng Yi; Lee, San Kan; Chang, Chein I.

In: Chinese Journal of Radiology, Vol. 31, No. 6, 01.12.2006, p. 275-287.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Mass detection in mammography using principle component analysis and stepwise selection

AU - Liao, Ping Sung

AU - Guo, Shu-Mei

AU - Yu, Nan Sue

AU - Chen, Cheng Yi

AU - Lee, San Kan

AU - Chang, Chein I.

PY - 2006/12/1

Y1 - 2006/12/1

N2 - This paper aims at constructing a feasible computer-aided mass detection system in mammography. To effectively distinguish mass tissues from normal tissues, the tissue characteristics are represented by a set of features, including fractal dimension, compactness, gray level histogram moments, statistics of spatial gray level dependence, texture spectrum and texture feature coding method, respectively. With reference to the energy viewpoint of principle component analysis (PCA), we explore possible linear combinations of the raw 212 feature variables, or from the original feature space we will extract several feature variables whose data distribution is complete enough to model that of the original feature space. Experimental results show that PCA is a useful tool to restrict the size of the independent feature variables, and stepwise selection approach in conjunction with PCA can sieve effective feature variables out from the raw feature space. In particular, on applying probabilistic neural network on the filtered feature vectors it has achieved reasonably good performance. The correct classification rates of fatty tissue, fatty-glandular and dense-glandular are 98.4%, 89.6% and 87.8%, respectively.

AB - This paper aims at constructing a feasible computer-aided mass detection system in mammography. To effectively distinguish mass tissues from normal tissues, the tissue characteristics are represented by a set of features, including fractal dimension, compactness, gray level histogram moments, statistics of spatial gray level dependence, texture spectrum and texture feature coding method, respectively. With reference to the energy viewpoint of principle component analysis (PCA), we explore possible linear combinations of the raw 212 feature variables, or from the original feature space we will extract several feature variables whose data distribution is complete enough to model that of the original feature space. Experimental results show that PCA is a useful tool to restrict the size of the independent feature variables, and stepwise selection approach in conjunction with PCA can sieve effective feature variables out from the raw feature space. In particular, on applying probabilistic neural network on the filtered feature vectors it has achieved reasonably good performance. The correct classification rates of fatty tissue, fatty-glandular and dense-glandular are 98.4%, 89.6% and 87.8%, respectively.

UR - http://www.scopus.com/inward/record.url?scp=34447509888&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34447509888&partnerID=8YFLogxK

M3 - Article

VL - 31

SP - 275

EP - 287

JO - Chinese Journal of Radiology

JF - Chinese Journal of Radiology

SN - 1018-8940

IS - 6

ER -