RSNA 2009 

Abstract Archives of the RSNA, 2009


SSM01-04

Comparison of Three Methods of Estimating Breast Density: BI-RADS Density Scores Using Full Field Digital Mammography, BI-RADS Density Scores Using Breast Tomosynthesis, and Volumetric Breast Density

Scientific Papers

Presented on December 2, 2009
Presented as part of SSM01: Breast Imaging (Digital Mammography)

Participants

Elizabeth Ann Rafferty MD, Presenter: Research grant, Hologic, Inc
Andrew Paul Smith PhD, Abstract Co-Author: Employee, Hologic, Inc
Loren Thomas Niklason PhD, Abstract Co-Author: Employee, Hologic, Inc Patent holder, Hologic, Inc Patent holder, Breast Tomosynthesis

PURPOSE

Compare three methods of estimating breast density. These methods included BI-RADS density scores for full field digital mammography (FFDM) and breast tomosynthesis. In addition, volumetric density based on x-ray transmission of the breast was calculated.

METHOD AND MATERIALS

Breast density of 310 subjects was scored by 15 radiologists. The radiologists scored the breast density using the BI-RADS breast density scale for both FFDM (Hologic Selenia FFDM system) and breast tomosynthesis (Hologic tomosynthesis prototype). For 264 cases the volumetric breast density was also measured using x-ray transmission measurements (Hologic R2 Quantra system). Breast composition was categorized as primarily fatty or dense by each method for each case and the results compared.

RESULTS

The average BI-RADS density scores using FFDM and tomosynthesis were nearly identical. Significant inter-reader variability was observed for both methods. The average standard deviation of radiologists’ BI-RADS scores for individual cases was 0.45 and 0.48 for FFDM and Tomosynthesis, respectively. Therefore the estimation and variability of breast density scoring was nearly identical for FFDM and tomosynthesis. A comparison of volumetric density to radiologists’ BI-RADS scoring demonstrated agreement in 83% of cases scored as dense and in 87% of cases scored as fatty.

CONCLUSION

Increased breast density has been shown to be associated with increased risk of breast cancer and decreased sensitivity of conventional mammography. Currently the standard method of estimating breast density is the BI-RADS score. This study provides evidence that radiologists’ estimates of breast density scored by the BI-RADS system are similar with FFDM and tomosynthesis, although significant inter-reader variability remains. Computer-based categorization of density also yielded similar density scores while offering the advantage of standardized scoring.

CLINICAL RELEVANCE/APPLICATION

Tomosynthesis and FFDM provide a similar estimation of breast density; volumetric breast density may provide an alternative method of determining breast density.

Cite This Abstract

Rafferty, E, Smith, A, Niklason, L, Comparison of Three Methods of Estimating Breast Density: BI-RADS Density Scores Using Full Field Digital Mammography, BI-RADS Density Scores Using Breast Tomosynthesis, and Volumetric Breast Density.  Radiological Society of North America 2009 Scientific Assembly and Annual Meeting, November 29 - December 4, 2009 ,Chicago IL. http://archive.rsna.org/2009/8016645.html