Room: Karl Dean Ballroom A1
Purpose: To test the validity of the assumption that the spectral content of the light emitted from a plastic scintillation detector (PSD) and its stem are invariant under arbitrary radiation field conditions, with emphasis on implications for surface dosimetry.
Methods: The detector consisted of Kuraray plastic scintillators coupled to an Eska optical fiber and a spectrometer. The PSDs were either oriented in the plane of gantry rotation or perpendicular to it. Gantry angle was varied between 0â?° and 45â?°. Normalized emission spectra of the scintillators and the fiber were compared for different orientations and gantry angles. Scintillation-only reference spectra were acquired on an orthovoltage x-ray unit. Stem spectra were acquired following scintillation measurements by removing the scintillator and repeating the measurements. Photons (6MV) and electrons (6 and 16MeV) were studied.
Results: Spectra acquired perpendicular to the gantry plane were invariant. Measurements in the plane of gantry rotation, were not invariant. Normalized stem spectra vary due to differences in the relative contributions of fluorescence and Cerenkov. The data suggest that Cerenkov produced in the PSD element is responsible for variations in the normalized scintillation spectrum. At a gantry angle of 45°, with 16 MeV electrons at the surface, we observed a 9.3% change in the peak amplitude of the normalized spectrum collected from the scintillator.
Conclusion: This work indicates that under certain conditions encountered in surface (i.e. in-vivo) dosimetry, assumptions about the invariance of the spectral content of the collected light signal are not valid. Errors in quantitative dose measurements depend on energy, radiation, PSD type and orientation, and depth. We observed a maximum error of 10% for 16 MeV electrons on the surface at a gantry angle of 45° when a stem subtraction method is used. Full spectral regression methods were accurate to within 2% under all conditions.