|
KMID : 1147120140200010053
|
|
Journal of the Korean Society of Imaging Informatics in Medicine
2014 Volume.20 No. 1 p.53 ~ p.57
|
|
|
Nipple Detection Using Image Processing on Digital Mammograms
|
|
Yoon Woong-Bae
Oh Ji-Eun
Kim Kwang-Gi
|
|
|
Abstract
|
|
|
|
Currently, mammography is the standard in breast screening and diagnosis. CAD(Computer aided diagnosis) technology that can improve the diagnostic accuracy of mammography has been continuously research and development. Segmentation is the process of partitioning images into regions, aiming to produce an image that is more meaningful and easier to analyze. Lesion location and image registration can be used to diagnose based on a nipple detection method. In this paper, we propose a method for detecting the nipple from the digital mammography. The proposed method detected the nipple using the brightness and the location information of outline. The auto nipple detection rate was 99%(139/140). A distance difference from the manually nipple point was minimum 0.2 mm up to 12.6 mm, average 2.6 mm. The detection rate of the nipple detected by the distance difference of less than 10mm is 97% (136/140). The proposed method is expected to be useful in the future with respect to breast cancer CAD.
|
|
|
KEYWORD
|
|
|
Nipple detection, Mammography, CAD
|
|
|
FullTexts / Linksout information
|
|
|
|
|
|
Listed journal information
|
|
|
|