Evaluation of the accuracy of an in vivo 3D dose reconstruction method using on-treatment EPID images
WenZhao Sun1, HongDong Liu1, SiJuan Huang1
1State Key Laboratory of Oncology in South China, Collaborative Innovation Center for Cancer Medicine, Sun Yat-sen University Cancer Center, Guangzhou, China,510060.
Purpose:
To validate a novel in vivo 3D dose reconstruction system through comprehensive comparisons with phantom measurements and TPS calculations.
Methods:
Validation was performed using homogeneous and anthropomorphic phantoms across multiple techniques and sites. Point and 2D doses were compared against measurements (with ion chamber and EBT3 film) and TPS calculations in homogeneous phantom. 3D dose validation were conducted both in homogeneous and anthropomorphic phantoms via gamma analysis (2 %-5%/2-3 mm). Sensitivity to morphological, setup, and delivery errors was evaluated using anthropomorphic phantoms.
Results:
The system demonstrated high dosimetric accuracy, with point dose differences between reconstructed, measured, and TPS-calculated values all within ± 1 %. Gamma passing rates (GPR) for planar dose comparisons exceeded 98 % (3 %/2 mm) and 94 % (2 %/2 mm). In 3D dose validation, GPR values exceeded 98 % in homogeneous phantoms and 97 % in anthropomorphic phantoms, even under the most stringent criterion (2 %/2 mm). For sensitivity and specificity, 3DCRT plans ranged from 1.00 and 0.75 (2 %/2 mm) to 1.00 and 1.00 (5 %/2 mm), while VMAT plans ranged from 1.00 and 1.00 (2 %/2 mm) to 0.667 and 1.00 (5 %/2 mm).
Conclusions:
A novel 3D dose reconstruction algorithm was developed to determine on-treatment 3D dose distributions in patients using transit EPID images. The system demonstrated high consistency with both measured doses and TPS-calculated doses, indicating its suitability for clinical implementation.
