KMID : 1147120090150010045
|
|
Journal of the Korean Society of Imaging Informatics in Medicine 2009 Volume.15 No. 1 p.45 ~ p.51
|
|
Automatic Breast Density Estimation Based on Statistical Image Information
|
|
Kim Young-Woo
Kim Chang-Won Kim Jong-Hyo
|
|
Abstract
|
|
|
This paper presents a new method for automatic breast density estimation based on statistical image information of a breast region in a mammogram. Breast density has been found to be a strong indicator for breast cancer risk, but measures of breast density still rely merely on a qualitative judgment of the radiologist. Therefore, objective and quantitative measurement is necessary to derive the relation between breast density and cancer risk. In this paper, we first perform a pre-processing to extract two features: statistical and edge information. Statistical information are mean and standard deviation of hypothetical fat and dense region in breast area. Edge information is the average of gradient magnitudes from a set of pixels having same intensity. These features are calculated over all pixel intensity range. By combining these two features, the optimal threshold is determined which best divides the fat and dense region. For evaluation purpose, a dataset of 80 cases of Full-Field Digital Mammography (FFDM) is utilized. Two human observers conducted a performance evaluation of proposed method. The correlation coefficients of the optimal threshold and estimated breast density between human observer and estimation were 0.958 and 0.987 on average, respectively. The experimental result indicates that the combination of statistic and boundary information is a novel method for automatic breast density estimation.
|
|
KEYWORD
|
|
breast density, CAD, image segmentation, mammography, FFDM
|
|
FullTexts / Linksout information
|
|
|
|
Listed journal information
|
|
|