Room: Exhibit Hall | Forum 9
Purpose: The purpose of this study is to quantify the three-dimensional deviation of localization for pancreatic treatments if matching were to be performed on bone instead of soft-tissue. This study helps evaluate utility of daily CBCT for SBRT treatments of pancreatic tumors.
Methods: Daily CBCT for SBRT delivery of eight pancreas patients were retrospectively evaluated for tumor localization based on soft-tissue and bony anatomy. All patients were treated with hypofractionation in 5 fractions. We calculated the means and standard deviations (SD) of the localization shifts based on bony anatomy using soft tissue alignment as the accepted standard. The systematic (Î£) and random (Ïƒ) errors for each localization were also calculated by the van Herkâ€™s method (2.5 Î£ + 0.7 Ïƒ).
Results: 40 CBCT sessions for eight patients were analyzed for preliminary results. The mean values and standard deviations for the localization shifts to regional bone anatomy from soft tissue were found to be 0.23 Â± 0.65 cm, 0.14 Â± 0.53 cm, and -0.11 Â± 0. 48 cm in the X, Y, and Z centroids, respectively. The three-dimensional deviation was calculated to be 1.25 Â± 1.10. The systematic errors based on the van Herkâ€™s method were calculated to be 0.65 cm, 0.53 cm, and 0.48 cm in the X, Y, and Z centroids, respectively. The random errors were determined to be 0.75 cm, 0.65 cm, and 1.15 cm in the X, Y, and Z centroids, respectively.
Conclusion: The translational random errors calculated are larger than the systematic errors in each centroid. This result indicates that the random errors have a bigger effect in the inaccuracy of treatments based on bone anatomy than the systematic errors. This study also shows integral role of CBCT and soft tissue based alignment during the SBRT treatments to the pancreatic tumor.