DOSE DISTRIBUTION IN A BREAST UNDERGOING MAMMOGRAPHY BASED ON A 3D DETAILED BREAST MODEL FOR CHINESE WOMEN

Dosimetry in a mammography phantom using TLD-300 dosimeters

PURPOSE

The purpose of this study was to evaluate the photon field effective energy (Eeff ) distribution and percentage depth-dose (PDD) within a mammography phantom by the analysis of the CaF2 :Tm (TLD-300) thermoluminescent (TL) glow curve. The experimental procedure involves the use of TLD-300 to determine with single dosimeter exposures both the relative dose and the beam quality.

METHODS

TLD-300 chips were exposed to x rays from a GE Senographe 2000D mammography unit at the surface and different depths within a BR12 phantom. X-ray beams were generated with Mo/Mo, Mo/Rh, and Rh/Rh anode/filter combinations and voltages between 25 and 34 kV. Glow curves were deconvoluted into component peaks and the high- to low-temperature ratio (HLTR) was evaluated. The photon field Eeff was obtained from the HLTR values using a calibration curve determined previously. PDD was established from the peak 5 TL signal (TLSP5 ) at depths between 0.0 and 3.5 cm inside the phantom. Taking into account the differences in density and composition between CaF2 :Tm and breast tissue, an energy-dependent correction factor (β(E)) was applied to TLSP5 . Measurements were compared with radiation transport Monte Carlo (MC) simulations performed with PENELOPE-2008.

RESULTS

A typical 5% change in the HLTR from the phantom top surface to 3.5 cm depth was measured, which corresponds to a 2.2 keV increase in photon field Eeff . Values of the β(E) correction factor were 0.33 and 0.13 for Eeff equal to 15.1 and 22.5 keV, respectively. This strong energy dependence of β(E) is mostly due to the differences in fluence attenuation between CaF2 and breast tissue. According to PDD measurements, dose decreased to half the surface value at depths between 0.7 and 1.0 cm for Mo/Mo/25 and Rh/Rh/34 beams, respectively. Values of PDD, less than 10% at 3.5 cm depth, would have been overestimated by about 3.5% (a large relative error) if an energy-independent correction factor had been assumed. Mean differences between experiments and MC simulations were 0.8 keV and 1.2% in the determination of Eeff and PDD, respectively.

CONCLUSION

The TLD-300 glow curve was used to accurately measure the photon field Eeff and PDD within a mammographic phantom. This work has demonstrated that Eeff and dose can be established simultaneously by using solely TLD-300.