Impact of Region-of-Interest Delineation Methods, Reconstruction Algorithms, and Intra- and Inter-Operator Variability on Internal Dosimetry Estimates Using PET.

PURPOSE: Human dosimetry studies play a central role in radioligand development for positron emission tomography (PET). Drawing regions of interest (ROIs) on the PET images is used to measure the dose in each organ. In the study aspects related to ROI delineation methods were evaluated for two radioligands of different biodistribution (intestinal vs urinary). PROCEDURES: PET images were simulated from a human voxel-based phantom. Several ROI delineation methods were tested: antero-posterior projections (AP), 3D sub-samples of the organs (S), and a 3D volume covering the whole-organ (W). Inter- and intra-operator variability ROI drawing was evaluated by using human data. RESULTS: The effective dose estimates using S and W methods were comparable to the true values. AP methods overestimated (49 %) the dose for the radioligand with intestinal biodistribution. Moreover, the AP method showed the highest inter-operator variability: 11 ± 1 %. CONCLUSIONS: The sub-sampled organ method showed the best balance between quantitative accuracy and inter- and intra-operator variability.

SU-E-T-290: Dosimetric Verification of Helical Tomotherapy against Experimental Measurements for Head and Neck Treatments.

PURPOSE: To verify a Tomotherapy plan for a typical head and neck treatment against experimental measurements. METHODS: The treatment plan for a head and neck case was generated by the Tomotherapy treatment planning system (TPS) to deliver ∼70 Gy in 33 sessions to the contoured PTV. The plan was calculated on a CIRS ATOM anthropomorphic phantom that provides a grid spacing of 3×3 cm2 holes to accommodate thermoluminescent detectors (TLD). The plan was verified against experimental measurements carried out by 7 LiF:Mg,Ti (TLD-700) TLD. Up to 20 locations were selected within the irradiated region and three detectors were used simultaneously at each point to decrease the statistical uncertainty. TLD locations were labeled in the planning system and dose comparisons between TPS prediction and experimental measurements were performed in terms of absolute dose to water for a single fraction. We examined the dose from (i) the corresponding 3.5MV Tomo-scan alone and (ii) the complete treatment. TLD-700 were found to fulfill the requirements of reproducibility, linearity and flat energy response in a previous study. In particular, TLD energy response was previously checked for 6 MV flattening filter free and conventional radiation beams under reference conditions. RESULTS: Doses derived from the TPS were in most cases in good agreement (4% on average) with TLD dose measurements within TLD statistical uncertainties (about 3%). Larger discrepancies up to 7% were found for points close to complex tissue inhomogeneities, such as bony structures. Dose from the scanning procedure alone is about 1 % of the dose per fraction. CONCLUSIONS: This work indicates that dose delivery plans created with Tomotherapy TPS are accurate for head and neck tumor localizations.