Monte Carlo evaluation of glandular dose estimates in X-ray breast - PowerPoint PPT Presentation

International Conference on Monte Carlo Techniques for Medical Application (MCMA2017) - Napoli 15 th -18 th October 2017 Monte Carlo evaluation of glandular dose estimates in X-ray breast computed tomography G. Mettivier, K. Bliznakova, A. Sarno , F. Di Lillo, J.M. Boone and P. Russo Napoli, 18 th October 2017 sarno@na.infn.it 1

Computed tomography dedicated to the breast - Fully 3D images - Uncompressed breast - 49-80 kVp W spectra J. M. Boone et al. Dedicated Breast CT: Radiation Dose and Image Quality Evaluation1. Radiology 221(3), 2001 2

Dosimetric parameters in breast CT MGD = DgN CT × K Air kerma at scanner isocenter Breast CT scanner at UC Davis Normalized glandular dose coefficient in CT calculated via MC simulations - Boone JM et al 2004 Med. Phys. - Thacker and Glick 2004 Phys. Med. Biol. - Sechopoulos et al 2010 Med. Phys. 3

Breast model and irradiation geometry - Cylindrical breast - Breast height = 1, 1.5, 2*breast radius - Homogeneous adipose/glandular mix - Skin thickness = 1.45 mm - Chest-to-central beam = 0 cm 4

Skin thickness influence on MGD 1.0 DgN CT deviation (unitary) 2 mm 0.8 3 mm 4 mm 0.6 to 1.45 mm skin thickness 0.4 Breast diameter = 14 cm Glandular fraction = 20% 0.2 10 20 30 40 50 60 Incident photon energy (keV) 5

Monoenergetic and Polyenergetic DgN CT 1.0 Monochromatic DgN 0.8 DgN CT (mGy/mGy) 0.55 0.6 Polychromatic DgN Glandular fraction: 0.50 0% pDgN CT (mGy/mGy) 25% 0.4 50% 0.45 75% 100% Breast diameter = 12 cm 0.40 0.2 Glandular fraction = 25% 0.35 0.0 10 20 30 40 50 60 0.30 49 kVp Incident photon energy (keV) HVL = 1.29 mm Al 0.25 6 8 10 12 14 16 18 Breast diameter (cm) 6

Dataset for monoenergetic DgN CT 1.0 Breast height= Radius x 2 Radius x 1.5 0.8 Radius x 1 DgN CT 0.6 2 > 0.9999 Fitting R 0.4 Breast diameter = 120 mm Breast glandularity = 25% 0.2 Calculated values 8th-order polynomlial fit-curve 0.0 10 20 30 40 50 60 70 80 Primary photon energy (keV) 7

Patient specific breast phantoms 8

Simple model vs patient specific breast phantoms A case study In this specific case the MGD calculated with the homogeneous cylindrical model is 21% lower than that calculated with the patient specific phantom (49 kVp; W/Al) 9

DgN CT coefficients validation 20 segmented 3D breasts 1.4 Hom. cont./Real breast model Max 49 kVp HVL = 1.29 mm Al Mean Std Min Max 1.2 th percentile 90 th percentile 75 Median Mean Glandular 1.0 th percentile 28.0 22.6 4.9 76.0 25 fraction (%) th percentile 10 0.8 Min Diameter 11.2 2.1 6.4 14.6 (cm) DgN 10 *results presented at ECMP 2016 by A. Sarno

Conclusions  A breast model for MGD evaluation in breast CT has been presented;  Monochromatic and polychromatic DgN coefficients have been provided;  The dose estimates with a simple model led to MGD differences when compared to that evaluated with patient specific breast phantoms;  The ongoing work for patient specific dose estimates has been shown. 11

Thank you!!! Any questions? International Conference on Monte Carlo Techniques for Medical Application (MCMA2017) - Napoli 15 th -18 th October 2017 sarno@na.infn.it 12