ESTRO 2024 - Abstract Book

S4831

Physics - Quality assurance and auditing

ESTRO 2024

The study was assessed in CIRS lung phantom. Elekta Sinergy linac equipped with Iview GT 3.4 EPID, and the TPS Monaco 5.12 (Elekta, Crawley, UK) have been used in this work. In vivo doses have been reconstructed by Dosimetry check 5.4 with collapsed cone convolution. Following the ESTRO protocol, three classes of errors have been directly implemented: setup errors (simulated by displacing the phantom 5-10-15 mm away from the isocenter); anatomical variations (simulated by placing 1-2-3 cm bolus on the phantom), delivery errors (simulated by rotating the collimator angle of 5-10-15 degrees). All errors were applied to a total of 9 plans: 6 3DCRT and 3 VMAT plans. Each error was also simulated into the TPS system, and the TPS results were used as gold standard for the presence of a clinical relevant error. The clinical relevant errors were defined as those errors causing a dose difference in the PTV of ΔD 50 > 5%, or ΔD 98 >10%, or ΔD 2 >10%. Different gamma metrics and Dose Volume Histogram (DVH) difference in the PTV have been computed (2%2mm, 3%2mm, 5%2mm, 2D and 3D, ΔD 50 , ΔD 98 , ΔD 2 in the PTV). ROC curves have been computed for each gamma metric, sensitivity and specificity of DVH metric have been also computed.

Results:

The sensitivities and specificities, for each metric and for plan type, have been reported in figure. 3D CRT plans showed higher sensitivity and a slightly lower specificity to error detection. On the contrary, the VMAT plans exhibit a lower sensitivity, and a good specificity. For 3D CRT plans, the highest Sensitivity/Specificity was 0.92/0.78, obtained with the 2D gamma 5% 2mm. For the VMAT plans better results have been obtained with the DVH metrics: showing Sensitivity/Specificity of 0.5/0.82.

Conclusion:

In this study we showed how to optimize the comparison metric for error detection of EPID IVD in lung treatment. Conformal and VMAT plans showed different optimal comparison metric, highlighting the importance of the plan dose conformity and the irradiation geometry for errors determination with EPID IVD.

Keywords: EPID in-vivo verification

References: