Validation of the Advanced Collapsed Cone Engine for Lead Shields In Contact Brachytherapy
Abstract
Purpose
To validate the model-based dose calculation algorithm (MBDCA) ACE against Monte Carlo (MC) for contact brachytherapy of the skin with the Freiburg flap applicator and lead shields.
Methods
Virtual phantoms of 2 nose treatments were designed to mimic real patient cases. The phantoms consist of virtual CTs with a voxel size of 1 mm3. Each voxel was assigned a tag number representing the voxel material. Treatment planning, dose calculations, and plan evaluations were performed in Oncentra Brachy, using the 192Ir TG186 generic source. The built-in HU thresholding tool was used for all structure delineations and included body, CTV, Freiburg flap, eyes, and a 3 mm thick lead shield, placed over the phantom’s eyes. Each structure was assigned a material with uniform density. Dose distributions were calculated with TG-43 and ACE. The MC toolkit TOPAS was used as a reference and validated according to the AAPM TG-186 recommendations, using test cases 1, 2, 3, and the interstitial breast case. The dose reporting grids were set to match the CT grid for all 3 algorithms. Selected dosimetric indexes and DVHs were compared in Oncentra Brachy.
Results
Evaluated dosimetric indexes in the CTV differed by less than 2 % between ACE and MC, whereas TG-43 showed a general dose overestimation of up to 7% for V100%. The dose reduction behind the 3 mm lead shields was correctly modeled by ACE, both D0.1cc and Dmean agreed within 1% to TOPAS.
Conclusion
Preliminary results indicate that the MBDCA ACE can be used to model the dose reduction behind 3 mm lead shields used in contact brachytherapy to reduce dose to normal tissues. The need for MBDCAs that go beyond the TG-43 formalism is evident for such cases and has the potential to be an important tool in treatment planning and clinical decision making.