Purpose: To validate a fully 3D-printed, heterogeneous mouse phantom for use in the dosimetric verification of radiobiological experiments.
Methods: An anatomically realistic mouse phantom was 3D printed based on segmented microCT data of a tumor-bearing 32-g mouse. Use of material jetting enabled material and color specification for each organ of interest. Various treatment configurations, delivered on a SARRP system, were used to validate dose delivery to both laser-cut EBT3 GafchromicÂ® films, located near the mouseâ€™s mid-plane, and a plastic scintillator dosimeter (PSD) that could be placed within target sites which included the brain, abdomen, or a 1-cm flank subcutaneous tumor. The PSD irradiations comprised a 3x3mmÂ² brain arc, 5x5mm2Â² POP and 5-beam 10x10mmÂ² abdominal co-planar arrangement. On-board CBCT targeting was used to localize the PSDâ€™s 1-mm active element prior to delivery. 2-D film dose distributions were acquired for a 3x3mmÂ² arc and 3-beam co-planar delivery using both the 5x5mmÂ² and 10x10mmÂ² fields. For each treatment configuration, dose to water was calculated with the TOPAS Monte Carlo (MC) code using a validated model of the SARRP and CAD geometries for the mouse phantom. Experimental and MC-derived film data were co-registered to enable 2-D dose comparison.
Results: Measured doses to the EBT3 film and scintillator dosimeters were found to agree to within 5% of the MC-calculated results at all target sites. PSD and MC-calculated dose differences were 1.33%, 1.51% and 3.17% for the 3x3,5x5 and 10x10mmÂ² fields, respectively. Mean differences between the film and MC central dose profiles for the 3x3,5x5 and 10x10mmÂ² irradiations were found to be 4.20%, 2.71% and 2.86%, respectively.
Conclusion: A fully 3D-printed mouse phantom was dosimetrically characterized and its corresponding Monte Carlo model validated. Using such a cost-effective phantom has been demonstrated as a valuable means to improve the reliability of pre-treatment dose verification in preclinical contexts.
Funding Support, Disclosures, and Conflict of Interest: This work was partially funded by the National Sciences and Engineering Research Council of Canada (NSERC) Discovery Grant