Evaluation of Setup Uncertainties for Single-Fraction SRS by Comparing the Two Different Mask-Creation Methods

The purpose of this study was to evaluate the setup uncertainties for single-fraction stereotactic radiosurgery (SF-SRS) based on the clinical data with the two different mask-creation methods using pretreatment CBCT imaging guidance. Dedicated frameless fixation BrainLAB masks for 23 patients were created as a routine mask (R-mask) making method, as explained in the BrainLAB user manual. The alternative masks (A-mask) which were created by modifying the cover range of the R-mask for the patient head were used for 23 patients. The systematic errors including the each mask and stereotactic target localizer were analyzed and the errors were calculated as the mean and standard deviation (SD) from the LR, SI, AP, and yaw setup corrections. In addition, the frequency of three-dimensional (3D) vector length were also analyzed. The values of the mean setup corrections for the R-mask in all directions were small; < 0.7 mm and < 0.1 degree, whereas the magnitudes of the SDs were relatively large compared to the mean values. In contrast to the R-mask, the means and SDs of the A-mask were smaller than those for the R-mask with the exception of the SD in the AP direction. The mean and SD in the yaw rotational direction in the R-mask and A-mask system were comparable. The 3D vector shifts of a larger magnitude occurred more frequently for the R-mask than the A-mask. The setup uncertainties for each mask with the stereotactic localizing system had an asymmetric offset towards the positive AP direction. The A-mask-creation method, which is capable of covering the top of the patient head is superior to that for R-mask, and thereby the use of the A-mask is encouraged for SF-SRS to reduce the setup uncertainties. Moreover, the careful mask making is required to prevent the possible setup uncertainties.