QUANTRA

Quantra™ is a software application designed to objectively assess breast tissue composition by calculating volumetric and area breast density from digital mammograms, utilizing an updated algorithm for improved consistency across acquisition parameters. It aims to provide radiologists with adjunctive, quantitative data to aid in standardized breast density reporting, including BI-RADS categorization, validated through phantom studies, reproducibility tests, and comparisons with reader assessments.

Quantra™ is a software application designed to work with images from digital breast x-ray systems to aid radiologists in assessing breast tissue composition. Its primary function is to calculate both volumetric breast density, as a ratio of fibroglandular tissue to total breast volume estimates, and area breast density, as a ratio of fibroglandular tissue area to total breast area estimates. The software also segregates breast density into categories, which can assist in reporting consistent BI-RADS® breast composition categories, particularly where mandated by state regulations. It's important to note that Quantra provides adjunctive information and is not intended as a direct interpretive or diagnostic tool. The Quantra 2.1 software operates on a Windows server and processes digital breast x-ray images. This version features an updated algorithm independent of certain x-ray acquisition parameters like spectrum and tube output, and incorporates a new reference population from Hologic Selenia Dimensions image sets. The method for combining image-level results for per-breast and per-patient density has been changed from using the maximum value to averaging. Performance testing included assessment of fibroglandular tissue thickness accuracy using a breast phantom, evaluation of breast density measurement reproducibility across different acquisition systems, statistical comparison of density measures between different views and breasts using a large multi-vendor image database, and comparison of its density measures against BI-RADS scores assigned by 15 readers. A risk analysis identified potential hazards, which are controlled via software development, verification, and validation testing, with the overall level of concern for these hazards classified as "Moderate".

AI Regulatory Concerns

Powered by

Jev reads the full FDA letter and returns calibrated concern probabilities per regulatory category. These are AI estimates, not regulatory advice.

• Q1. How does the updated algorithm in Quantra 2.1 achieve independence from x-ray spectrum and tube output acquisition parameters, and what was the impact of this change on density measurement consistency across different machines?; What were the specific quantitative results of the phantom study assessing the measurement accuracy for fibroglandular tissue thickness?; Could you elaborate on the statistical agreement found between Quantra's density measures (q_abd, Q_abd, Vbd, Abd) and the BI-RADS categories assigned by the 15 readers, and were there any specific BI-RADS categories where agreement was notably higher or lower?; How does the new method of averaging (versus maximum in Quantra 2.0) for combining image-level results affect the per-breast and per-patient breast density measures, particularly in cases of asymmetric density?; What are the computational requirements for the Quantra software, and how does it integrate into existing Picture Archiving and Communications Systems (PACS) and radiology workflows beyond operating on a Windows server?