Multi Domain Features Based Classification of Mammogram Images Using SVM and MLP

Breast cancer is the most common cancer diagnosed among U.S. women. In this paper we have done some experiments for tumor detection in digital mammogram images. First of all, we have described a method that segments the breast image automatically. As a preprocessing, we have used fuzzy based noise r...

Full description

Saved in:
Bibliographic Details
Published in2009 Fourth International Conference on Innovative Computing, Information and Control (ICICIC) pp. 1301 - 1304
Main Authors Jaffar, M.A., Ahmed, B., Hussain, A., Naveed, N., Jabeen, F., Mirza, A.M.
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.12.2009
Subjects
Breast cancer
Cancer detection
Computer science
Diseases
Enterprise resource planning
History
Image segmentation
Mammography
Support vector machine classification
Support vector machines
Online AccessGet full text

Cover

More Information
Summary:Breast cancer is the most common cancer diagnosed among U.S. women. In this paper we have done some experiments for tumor detection in digital mammogram images. First of all, we have described a method that segments the breast image automatically. As a preprocessing, we have used fuzzy based noise removal filter that removes noise. Then for segmentation, we have provided a background removal method. We have extracted eight different multi domains features. For accurate classification, we have used two different classification techniques: Support Vector Machine (SVM) and Multilayer Perceptrons (MLP). We have compared our results with a method that has used 8 features. We have shown results that four features are not sufficient for classification. Results show the superiority of the proposed algorithm in terms of sensitivity, specificity and accuracy. We have used MIAS [7] database of mammography.
ISBN:142445543X
9781424455430
DOI:10.1109/ICICIC.2009.270