Estimation of mean glandular dose using breast-equivalent phantoms: a foundational study for regional DRL formulation in mammography in India

The objective of this foundational study is to support the development of regional diagnostic reference levels (DRLs) for mammography screening in India by estimating mean glandular dose (MGD) using appropriate breast-equivalent phantoms across computed mammography (CR), digital mammography (DR), and digital breast tomosynthesis (DBT) systems. Additionally, system-displayed MGD values were compared with calculated MGD values to evaluate their accuracy for routine clinical use. A total of 5000 mammographic views were collected across CR, DR, and DBT units. For each unit and phantom thickness, ten entrance surface air kerma (ESAK) measurements were performed under clinical settings using a calibrated solid-state dosimeter (AccuGold+, Radcal, USA). Breast-equivalent slab phantoms (Sun Nuclear, USA), ranging from 30 mm to 90 mm, were used to simulate varying glandular compositions. Standard American College of Radiology and DBT phantoms were employed to ensure consistent imaging conditions. MGDs were derived from ESAK values using IAEA recommended conversion factors. For DR and DBT systems, calculated MGDs were compared with system displayed values. The 75th percentile MGDs for each phantom thickness were used to propose regional DRLs, in line with ICRP Publication 135 and IAEA Human Health Report No. 77. At the standard 60 mm compressed breast thickness, the 75th percentile MGDs were found to be 2.58 ± 0.38 mGy (CR), 1.68 ± 0.86 mGy (DR), and 2.66 ± 0.78 mGy (DBT). System-displayed MGDs in DR and DBT were observed to be 13%-17% higher than calculated values. This study presents a scalable, reproducible framework for establishing regional, thickness-specific DRLs in India, supporting dose optimisation and quality assurance in mammography practice.