Machine learning-based analysis of average glandular dose in mammography: The role of breast density.

AI / informaticsRadiography (London, England : 1995)· Aug 2026

Machine learning-based analysis of average glandular dose in mammography: The role of breast density.

A González-Ruíz, H I Sánchez Mendoza, C Ferreira Martinez … H Peregrina-Barreto

Read paper

Breast ImagingAI / informaticsQ2IF 2.6

Breast density is the second most important predictor of mammography radiation dose.

Machine learning-based analysis of average glandular dose in mammography: The role of breast density.

A González-Ruíz … H Peregrina-Barreto

Radiography (London, England · Aug 2026 · Q2 in Radiology

Background

Average glandular dose (D g ) is the primary metric for assessing radiation risk in screening mammography.

Purpose

This study evaluated the influence of breast density on D g variability using the TG282-based dosimetry model and a machine learning approach.

Methods

The cross-sectional study included 1233 full-field digital mammography images from three mammography units.

n = 1233 full-field digital mammography

Results

breast density was the second strongest predictor of dose variability, after compression force

45.72%

Compression force

23.93%

Breast density

More results

D g distributions were non-normal, and D g was higher in MLO than CC views (p < 0.05).

Significant differences were observed across compressed breast thickness and breast density categories.

D g correlated positively with mAs and exposure time, while breast density and compression force showed moderate associations.

More results

D g,75 benchmarks increased with breast density within compressed breast thickness groups.

“

Conclusion · 1 of 3

Incorporating breast density improves the characterisation of D g variability in mammography.

Conclusion · 2 of 3

The Random Forest model characterised multivariable relationships between breast density and compression-related factors, supporting more anatomically informed dose optimisation strategies.

Conclusion · 3 of 3

IMPLICATIONS FOR PRACTICE: breast density-stratified D g,75 benchmarks may improve identification of atypical dose values and support targeted dose monitoring in mammography.

Machine learning-based analysis of average glandular dose in mammography: The role of breast density.

Machine learning-based analysis of average glandular dose in mammography: The role of breast density.

Related papers