Comparative Evaluation of classifiers and Feature Selection Methods for Mass Screening in Digitized Mammograms

Author

Wang, Chuin-Mu ; Yang, Sheng-Chih ; Chung, Pau-Choo

Author_Institution

Dept. of Comput. Sci. & Inf. Eng., Nat. Chin Yi Inst. of Tech., Taichung

fYear

2006

fDate

38899

Firstpage

1

Lastpage

2

Abstract

In this paper, three groups of characteristics related to mass texture are adopted, namely, SGLD (spatial gray level dependence), TS (texture spectrum) and TFCM (texture feature coding method) to describe the characteristics of masses and normal textures on digitized mammograms. Next, under the testing by classifiers, three feature selection methods - SBS (sequential backward selection), SFS (sequential forward selection) and SFSM (sequential floating search method) are used to find out suboptimal subset from 19 features in order to improve the performance of mass detection. Finally, two classifiers PNN (probabilistic neural network) and SVM (support vector machine) are applied and their performances are compared

Keywords

image classification; image texture; mammography; medical image processing; neural nets; support vector machines; classifiers; digitized mammograms; feature selection methods; mass screening; mass texture; probabilistic neural network; sequential backward selection; sequential floating search method; sequential forward selection; spatial gray level dependence; support vector machine; texture feature coding method; texture spectrum; Breast; Computer science; Feature extraction; Image texture analysis; Neural networks; Pixel; Search methods; Sequential analysis; Support vector machine classification; Support vector machines;

fLanguage

English

Publisher

ieee

Conference_Titel

Life Science Systems and Applications Workshop, 2006. IEEE/NLM

Conference_Location

Bethesda, MD

Print_ISBN

1-4244-0277-8

Electronic_ISBN

1-4244-0278-6

Type

conf

DOI

10.1109/LSSA.2006.250418

Filename

4015819