Related Experiment Video
Updated: Jul 12, 2025
15:48
Tracking the Mammary Architectural Features and Detecting Breast Cancer with Magnetic Resonance Diffusion Tensor Imaging
Published on: December 15, 2014
22.5K
Breast density analysis of digital breast tomosynthesis
John Heine1, Erin E E Fowler2, R Jared Weinfurtner3
1Cancer Epidemiology Department, Moffitt Cancer Center and Research Institute, 12902 Bruce B. Downs Blvd, Tampa, FL, 33612, USA. john.heine@moffitt.org.
Scientific Reports
|November 1, 2023
Summary
Digital breast tomosynthesis (DBT) enables automated breast density (PD) assessment for breast cancer risk. Volumetric and synthetic 2D image measures showed significant associations with cancer risk, validating DBT applications.
Area of Science:
- Radiology and Medical Imaging
- Oncology
- Biostatistics
Background:
- Mammography has transitioned to digital breast tomosynthesis (DBT) in the United States.
- Automated percentage of breast density (PD) techniques, initially for 2D mammography, require evaluation with DBT data.
Purpose of the Study:
- To evaluate an automated PD technique with DBT for breast cancer risk prediction.
- To compare different PD measurement approaches derived from DBT images, including volumetric and synthetic 2D methods.
Main Methods:
- Assessed PD using normalized-volumetric, dense volume, slice-mean, and synthetic 2D image methods with DBT.
- Derived volumetric measures and modeled PD as a function of compressed breast thickness (CBT).
- Conducted a matched case-control study (n=426 pairs) to estimate odds ratios (ORs) for breast cancer risk.
Main Results:
- Significant odds ratios (ORs) were found for PD measures derived from DBT.
- Volumetric and slice-mean PD measures showed similar significant associations with risk (OR ≈ 1.43-1.44).
- PD modeled as a function of CBT (OR ≈ 1.47) and means from volume/synthetic images (OR ≈ 1.31) were also significant.
Conclusions:
- Automated PD assessment using DBT is effective for breast cancer risk prediction.
- Volumetric and synthetic 2D image analyses from DBT provide valuable risk information.
- The developed methods offer alternative approaches for standardized 2D synthetic image construction in DBT analysis.
