Room: Track 1
Purpose: To investigate energy deposition in glandular tissues across microscopic-to-macroscopic length scales for different advanced x-ray breast imaging modalities: mammography, digital breast tomosynthesis (DBT), contrast enhanced digital mammography (CEDM) and breast-CT (BCT).
Methods: We studied energy deposition in breast tissue with Monte Carlo simulations using an anthropomorphic, segmented breast phantom comprised of 0.5 mm³ voxels, 20% glandular tissue by mass, with 5 cm compressed thickness. First, the MCGPU code was used to score mean glandular dose (MGD), glandular voxel doses and phase space files within voxels for different breast imaging modalities. Afterwards, the phase space files were loaded in PENELOPE and populations of 5796 fibroglandular cells (each modelled as concentric spheres of radii 6 and 9 µm representing nucleus and cytoplasm) were irradiated, scoring the specific energy distributions over nucleus and cytoplasm targets.
Results: For a 4 mGy MGD, 3D dose distributions in voxels comprised of glandular tissues varied significantly over the breast volume and between different imaging modalities, with ranges (minimum-maximum doses) as follows: 0.52-16.2 mGy (mammography), 0.72-15.1 mGy (DBT), 1.54-6.22 mGy (CEDM), and 1.90-7.89 mGy (BCT). Nucleus average specific energies are 6.3% (cytoplasm: 5.7%) higher than the respective voxel glandular dose across the different imaging modalities. Doses and specific energies are highly sensitive to assumed breast tissue and cell elemental compositions. The fraction of cell nuclei receiving energy is 85.7% for a 1 mGy glandular voxel dose (100% for 11 mGy).
Conclusion: The energy deposition inside the breast varies considerably over micro- and macroscopic length scales and imaging modalities. The multiscale simulations open a new perspective to better understand breast dosimetry, providing additional tools to better estimate the associated risks of ionizing radiation and compare different imaging approaches.
Funding Support, Disclosures, and Conflict of Interest: Funding support from: Natural Sciences and Engineering Research Council (NSERC) of Canada; Ontario Ministry of Research and Innovation; Canada Research Chairs; Conselho Nacional de Desenvolvimento Cientifico e Tecnologico (CNPQ) project 140155/2019-8; FAPESP project 2015/21873-8; ELAP scholarship with support of the Government of Canada; AAPM International Training and Research Coordination scholarship.