Congress:
EuroSafe Imaging 2021
Keywords:
Breast, Contrast agents, Radioprotection / Radiation dose, Mammography, Contrast agent-intravenous, Dosimetry, Radiation safety, Cancer, Multidisciplinary cancer care, Quality assurance
Authors:
E. M. L. Vaccara, N. Brunetti, M. Calabrese, A. Garlaschi, L. Ridolfi, G. Pellecchia, M. Dordolo, C. Gavotti, F. Levrero
DOI:
10.26044/esi2021/ESI-13005
Methods or background
As is known, CEDM consists of two exposures:
- The first realized with low energy due to W/Rh or W/Ag (Anode/Filter combination)
- The second at high energy due to W/Cu.
The internal protocol provides for the injection of the iodinated contrast agent and the wait of two minutes; then, for each craniocaudal (CC) and mediolateral oblique (MLO) projection on both breasts. Typically, it begins from the side where there is the disease you want to investigate. The system acquires two separated but nearly simultaneous low and high energy exposures (for a total of 8 exposures per patient). The overall dose for each patient is the sum of the 8 contributions mentioned.
The acceptance test shows high image quality in both modalities, CEDM and FFDM, (TORMAX images, analysed with AutoPIA 3.7.7, reaches a Frequency at MTF 50% greater than 6 lp/mm).
To monitor first enrolled patients in CEDM from the radiation protection point of view, we compare the AGD calculated by the manufacturer with our AGD evaluated from Entrance Surface Air Kerma (ESAK) measured for several Poly Methyl MethAcrylate (PMMA) thicknesses. For the ESAK measurements, a newly calibrated Radcal 10x6-6M ionization chamber was used, with polystyrene support dedicated to mammography, to have an accurate positioning reproducibility.
First, we calculated the accuracy of the phantom AGD depending on the thickness for both the FFDM and the new CEDM technique.
Therefore CEDM AGD is the result of two AGD values each affected by its own accuracy error. As shown in the following table,
the agreement between the PMMA dose calculated by the Hologic and our estimates is very good for all image acquisition modes. The variability of the data, expressed by the standard deviation, is always less than 15%.