(Monday, 7/15/2019) 1:15 PM - 1:45 PM
Room: Exhibit Hall | Forum 6
Purpose: To evaluate the Advanced Convolution Engine (ACE) algorithm with a patient specific, low-cost, anthropomorphic phantom constructed via 3D-printing. Model-based inverse-planning algorithms are becoming increasingly available in brachytherapy, the need now arises for a means of appropriately testing their ability to accurately predict the behavior of radiation within heterogeneous materials.
Methods: The CT imaging of a patient during a treatment with ‘Venezia’ HDR instruments was used as a basis for the construction of a low-cost testing phantom. CT images were segmented by HU value with the open-source ‘3DSlicer’ software. CT voxel values nearest to water-equivalent were segmented to create 3D-printed phantom sections with a desktop-style 3D printer. The semi-hollow reconstructions of PLA plastic were filled with low-cost tissue equivalent materials corresponding to regions identifiable as bone, fat, or muscle. The phantom was CT imaged and planned upon with Elekta’s ACE and TG34 algorithms and the resulting dose distributions were compared to those obtained during patient treatment. A gamma analysis of calculated dose distributions and irradiated GafChromic EBT3 films were used to evaluate the dose predicted within the printed phantom by the TG43 and ACE algorithms.
Results: The gamma analysis of a series of patient treatment dose planes calculated with ACE and TG43 algorithms against films of plans delivered to patient-based phantom showed an improvement in accuracy for the ACE algorithm over those predicted by the TG43 algorithm.
Conclusion: Affordable 3D printing options and readily available tissue-equivalent materials can allow for the creation of custom anthropomorphic phantoms for any clinic with limited budgets that desire to test the viability of clinically implementing model-based planning for brachytherapy.
Keywords
Not Applicable / None Entered.
Taxonomy
Not Applicable / None Entered.
Contact Email
