|
KMID : 1114620120090020127
|
|
Journal of the Korean Society for Breast Screening
2012 Volume.9 No. 2 p.127 ~ p.132
|
|
|
Mammographic density assessment: Comparison of VolparaTM Software and Visual BI-RADS Classification
|
|
Choi Ji-Soo
Lee Chan-Wha
Suh Hee-Jung
Kim Min-Ju
Lee Ju-Hee
Lee Chang-Yoon
Joo Jung-Nam
Ko Kyoung-Ran
|
|
|
Abstract
|
|
|
Purpose: To evaluate the reproducibility of Breast Imaging Reporting and Data System (BI-RADS) classification by radiologists and to compare Volpara estimation with BI-RADS classification in mammographic density assessment.
Materials and Methods: From September to November 2011, 913 patients in whom mammography was performed at a referral center and mammographic density was graded by Volpara software were included in this study. Mammographic density was assessed by seven radiologists according to BI-RADS. Majority report of radiologists was used as standard reference to compare BI-RADS to Volpara classification for mammographic density assessment. Agreement between the two methods and the reference standard was assessed using kappa test according to four categories (D1/2/3/4) and two categories (D1-2/D3-4).
Results: Intraobserver (k range 0.69-1.00) agreement of the seven radiologists was substantial or almost perfect, and interobserver (k range 0.55-0.77) agreement of the radiologists was moderate or substantial. The agreement with the reference standard was substantial or almost perfect for BI-RADS classifications made by radiologists (D1/2/3/4: k range 0.75-0.86, D1-2/D3-4: k range 0.70-0.90), and was moderate for Volpara classification (D1/2/3/4: k = 0.55, D1-2/D3-4: k = 0.59).
Conclusion: Mammographic density classification by radiologists according to BI-RADS was highly reproducible. The agreement between BI-RADS and Volpara classifications was moderate in mammographic density estimation.
|
|
|
KEYWORD
|
|
|
Mammographic breast density, Volpara, Breast Imaging Reporting and Data System
|
|
|
FullTexts / Linksout information
|
|
|
|
|
|
Listed journal information
|
|
|
|