No abstract

The purpose of this study is to evaluate the prediction and classification performances of the gamma passing rate (GPR) for different machine learning models and to select the best model for achieving machine learningbased patient-specific quality assurance (PSQA). Methods: The measurement verification of 356 head-and-neck volumetric modulated arc therapy plans was performed using a diode array phantom (Delta4 Phantom), and GPR values at 2%/2 mm with global normalization and 3%/2 mm with local normalization were calculated. Machine learning models, including ridge regression (RIDGE), random forest (RF), support vector regression (SVR), and stacked generalization (STACKING), were used to predict the GPR. Each machine learning model was trained using 260 plans, and the prediction accuracy was evaluated using the remaining 96 plans. The prediction error between the measured and predicted GPR was evaluated. For the classification evaluation, the lower control limit for the measured GPR and lower control limit for predicted GPR (LCL p ) was defined to identify whether the GPR values represent a "pass" or a "fail." LCL p values with 99% and 99.9% confidence levels were calculated as the upper prediction limits for the GPR estimated from the linear regression between the measured and predicted GPR. Results: There was an overestimation trend of the low measured GPR. The maximum prediction errors for RIDGE, RF, SVR, and STACKING were 3.2%, 2.9%, 2.3%, and 2.2% at the global 2%/2 mm and 6.3%, 6.6%, 6.1%, and 5.5% at the local 3%/2 mm, respectively. In the global 2%/2 mm, the sensitivity was 100% for all the machine learning models except RIDGE when using 99% LCL p . The specificity was 76.1% for RIDGE, RF, and SVR and 66.3% for STACKING; however, the specificity decreased dramatically when 99.9% LCL p was used. In the local 3%/2 mm, however, only STACKING showed 100% sensitivity when using 99% LCL p . The decrease in the specificity using 99.9% LCL p was smaller than that in the global 2%/2 mm, and the specificity for RIDGE, RF, SVR, and STACKING was 61.3%, 61.3%, 72.0%, and 66.8%, respectively. Conclusions: STACKING had better prediction accuracy for low GPR values than other machine learning models. Applying LCL p to a regression model enabled the consistent evaluation of quantitative and qualitative GPR predictions. Adjusting the confidence level of the LCL p helped improve the balance between the sensitivity and specificity. We suggest that STACKING can assist the safe and efficient operation of PSQA.

To measure the absorbed dose to water D in proton beams using a radiophotoluminescent glass dosimeter (RGD), a method with the correction for the change of the mass stopping power ratio (SPR) and the linear energy transfer (LET) dependence of radiophotoluminescent efficiency [Formula: see text] is proposed. The calibration coefficient in terms of D for RGDs (GD-302M, Asahi Techno Glass) was obtained using a Co γ-ray. The SPR of water to the RGD was calculated by Monte Carlo simulation, and [Formula: see text] was investigated experimentally using a 70 MeV proton beam. For clinical usage, the residual range R was used as a quality index to determine the correction factor for the beam quality [Formula: see text] and the LET quenching effect of the RGD [Formula: see text]. The proposed method was evaluated by measuring D at different depths in a 200 MeV proton beam. For both non-modulated and modulated proton beams, [Formula: see text] decreases rapidly where R is less than 4 cm. The difference in [Formula: see text] between a non-modulated and a modulated proton beam is less than 0.5% for the R range from 0 cm to 22 cm. [Formula: see text] decreases rapidly at a LET range from 1 to 2 keV µm. In the evaluation experiments, D using RGDs, [Formula: see text] showed good agreement with that obtained using an ionization chamber and the relative difference was within 3% where R was larger than 1 cm. The uncertainty budget for [Formula: see text] in a proton beam was estimated to investigate the potential of RGD postal dosimetry in proton therapy. These results demonstrate the feasibility of RGD dosimetry in a therapeutic proton beam and the general versatility of the proposed method. In conclusion, the proposed methodology for RGDs in proton dosimetry is applicable where R > 1 cm and the RGD is feasible as a postal audit dosimeter for proton therapy.