Parenchymal breast density estimation with the use of statistical characteristics and textons

Chatzistergos, S; Stoitsis, J; Papaevangelou, A; Zografos, G; Nikita, KS

dc.contributor.author	Chatzistergos, S	en
dc.contributor.author	Stoitsis, J	en
dc.contributor.author	Papaevangelou, A	en
dc.contributor.author	Zografos, G	en
dc.contributor.author	Nikita, KS	en
dc.date.accessioned	2014-03-01T02:46:54Z
dc.date.available	2014-03-01T02:46:54Z
dc.date.issued	2010	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/32936
dc.subject	Density Estimation	en
dc.subject	Text Classification	en
dc.subject	Breast Cancer	en
dc.subject.other	Breast abnormalities	en
dc.subject.other	Breast cancer risk	en
dc.subject.other	Breast density	en
dc.subject.other	Breast density estimation	en
dc.subject.other	Breast tissues	en
dc.subject.other	CAD system	en
dc.subject.other	Classification rates	en
dc.subject.other	Mammographic images	en
dc.subject.other	Statistical characteristics	en
dc.subject.other	Text classification	en
dc.subject.other	Textons	en
dc.subject.other	Textural characteristic	en
dc.subject.other	Visual word	en
dc.subject.other	Information technology	en
dc.subject.other	Mammography	en
dc.subject.other	Text processing	en
dc.title	Parenchymal breast density estimation with the use of statistical characteristics and textons	en
heal.type	conferenceItem	en
heal.identifier.primary	10.1109/ITAB.2010.5687686	en
heal.identifier.secondary	http://dx.doi.org/10.1109/ITAB.2010.5687686	en
heal.identifier.secondary	5687686	en
heal.publicationDate	2010	en
heal.abstract	Breast parenchymal density has been found to be a strong indicator for breast cancer risk but the process of classification by the radiologists has been proved to be quite subjective. Furthermore, recent studies have shown that the effectiveness of most modern CAD systems used for breast abnormality detection diminishes greatly when parenchymal breast tissue density is high. Therefore the existence of a system to provide automatically accurate and objective estimation of the parenchymal breast density category is of great importance. In the present work we present a method to perform breast density classification based on local textural characteristics of the image, known as textons. We make the assumption that the mammographic image comprises a ""document"" with a number of ""visual words"" and the ""document"" ""topic"" is the wanted density category. This approach allows the use of techniques originating from text classification which have been proved to be very effective. The proposed system achieved classification rates of 98.3% and 93.2% for two categories and four categories, respectively. © 2010 IEEE.	en
heal.journalName	Proceedings of the IEEE/EMBS Region 8 International Conference on Information Technology Applications in Biomedicine, ITAB	en
dc.identifier.doi	10.1109/ITAB.2010.5687686	en