DocumentCode
2044851
Title
An improved non-linear transformation function for enhancement of mammographic breast masses
Author
Bhateja, Vikrant ; Devi, Swapna
Author_Institution
Dept. of Electron. & Commun. Eng., SRMCEM, Lucknow, India
Volume
5
fYear
2011
fDate
8-10 April 2011
Firstpage
341
Lastpage
346
Abstract
Breast cancer is often diagnosed at its early stage as a non-palpable suspicious mass on a mammogram that may be cancerous at a later stage. Masses can be seen in both mediolateral and craniocaudal mammographic views. Masses with diameters less than 1 cm are generally benign, while those with diameters above 2 cm require further investigation. This paper presents a novel approach for enhancement of suspicious masses in mammographic images using the proposed nonlinear transformation function. Significant improvement in contrast of masses along with the suppression of background tissues is obtained by tuning the parameters of the proposed transformation function in the specified range. The selection of parameters is almost invariant of the type of background tissues and severity of the abnormality, giving significantly improved results even for denser mammographic images. The applicability of the proposed method is tested on mammograms with circumscribed, ill-defined and spiculated masses.
Keywords
biological organs; cancer; mammography; medical image processing; background tissue suppression; breast cancer; craniocaudal mammographic view; mammographic breast mass enhancement; mediolateral mammographic view; nonlinear transformation function; Breast cancer; Image edge detection; Lesions; Pixel; Tuning; MIAS database; circumscribed masses; contrast enhancement measure (CEM); double sigmoid function; nonlinear transformation function; spiculated masses;
fLanguage
English
Publisher
ieee
Conference_Titel
Electronics Computer Technology (ICECT), 2011 3rd International Conference on
Conference_Location
Kanyakumari
Print_ISBN
978-1-4244-8678-6
Electronic_ISBN
978-1-4244-8679-3
Type
conf
DOI
10.1109/ICECTECH.2011.5942016
Filename
5942016
Link To Document