Computed Radiography (CR) マンモグラフィにおける乳腺含有率の推定

Translated title of the contribution: Computerized Estimation of a Percent Glandular Tissue Composition in Computed Radiography Mammography

辻田 直子, Sachiko Gotou, Yoshiharu Azuma, 白石 順二

Research output: Contribution to journalArticle

Abstract

Measurement of a percent glandular tissue composition (%GTC) is important in terms of the estimation of individual patient exposure dose and the prediction of malignancy, and thus a number of reports for estimating %GTC by use of a mammogram have been published. In this study, we propose a method for estimating individual %GTC by use of computed radiography (CR) mammograms. By employing breast-equivalent phantoms that are able to create breast phantom images with various combinations of fat and glandular tissue, as well as the thickness of whole breast, we determined a reference table for converting an each pixel value on CR mammography to the glandular tissue ratio. Therefore, the %GTC for individual breast was estimated by averaging glandular tissue ratio for a whole region. The clinical image data set that consisted of 49 CR mammograms were used for estimating %GTC. A paired comparison method for determining subjective ranking of the degree of breast density was employed in order to demonstrate the validity of our method. The results indicate that the average estimated %GTC was 35.0% (ranged from 12.0% to 67.0%) and they had a increased correlation with the ranking of those obtained by observer test. Therefore, it was suggested that our proposed method would be utilized for estimating the %GTC in objective manner.
Translated title of the contributionComputerized Estimation of a Percent Glandular Tissue Composition in Computed Radiography Mammography
Original languageJapanese
Pages (from-to)1540-1547
Number of pages8
JournalNippon Hoshasen Gijutsu Gakkai zasshi
Volume67
Issue number12
DOIs
Publication statusPublished - Dec 20 2011

Keywords

  • a percent glandular tissue composition
  • computed radiography (CR)
  • mammography
  • breast-equivalent phantom

Fingerprint Dive into the research topics of 'Computerized Estimation of a Percent Glandular Tissue Composition in Computed Radiography Mammography'. Together they form a unique fingerprint.

  • Cite this