Evaluation of breast parenchymal density with QUANTRA software.

Area of Science:

• Radiology
• Medical Imaging Analysis

Background:

• Breast density is a key factor in mammographic interpretation and breast cancer risk assessment.
• Current visual assessment of breast density using the American College of Radiology (ACR) Breast Imaging Reporting and Data System (BI-RADS) categories can be subjective.

Purpose of the Study:

• To evaluate the QUANTRA software for quantifying breast parenchymal density.
• To correlate the numerical breast density values from QUANTRA with the established ACR BI-RADS categories.

Main Methods:

• Digital mammograms from 545 women were visually categorized into ACR BI-RADS D1-D4 by three radiologists.
• QUANTRA software generated numerical breast density values.
• Receiver operator characteristic (ROC) analysis was used to determine cutoff values for each BI-RADS category.

Main Results:

• QUANTRA-derived breast density values were systematically lower than visual estimates.
• Specific QUANTRA value cutoffs accurately differentiated BI-RADS categories: <14.5% for D1 vs. D2-D4 (AUC 94.09%), <19.5% for D1-D2 vs. D3-D4 (AUC 94.4%), and <26.5% for D1-D3 vs. D4 (AUC 90.75%).

Conclusions:

• QUANTRA software can establish objective cutoff values for ACR BI-RADS breast density categories.
• Despite lower numerical values, QUANTRA density measurements show good correlation with visual BI-RADS assessments, enhancing standardization.