(Sunday, 7/29/2018) 3:00 PM - 6:00 PM
Room: Exhibit Hall
Purpose: The purpose of our study is to confirm the accuracy of a radiation isocenter analysis system (RIAS) in an Elekta linear accelerator (Linac).
Methods: (1) Mega voltage (MV) images of a 8-mm tungsten ball baring (BB) were scanned at a gantry angle of 0 degree after the BB was moved to four known distances (0.5, 1.0, 2.0, 3.0 mm). The acquired images were imported into a free image analysis software (ImageJ; IJ), and the deviation between the shift distance of the BB and the known shift amount was calculated. (2) MV images of a BB were scanned from four gantry angles (0, 90, 180, 270°) including two collimator angles (0, 180°) after the BB was moved to four known distances in the I/S, L/R and A/P directions. The IJ and RIAS were statistically compared for errors with respect to the known shift amount.
Results: (1) The means ± SD (max) of the measured shift errors by using IJ was 0.012 ± 0.009 (0.025), 0.014 ± 0.009 (0.026), 0.019 ± 0.015 (0.042), and 0.03 ± 0.017 (0.056) mm for the four known distances, respectively. (2) For the IJ, the means ± SD (max) of the measured shift deviations was 0.06 ± 0.04 (0.18), 0.2 ± 0.11 (0.43), and 0.02 ± 0.02 (0.09) mm in the S/I, L/R, and A/P directions, respectively. For RIAS, the means ± SD (max) was 0.13 ± 0.09 (0.4), 0.24 ± 0.15 (0.55), and 0.06 ± 0.04 (0.18) mm in the S/I, L/R, and A/P directions, respectively. The systematic errors from the I/S and A/P directions were statistically significant (PS/I = 0.00001 and PA/P = 0.00007 by Welch’s t-test).
Conclusion: RIAS in an Elekta linac provides good accuracy for determining the center of MV images. However, RIAS was slightly overestimated compared with the known shift amount.
Keywords
Not Applicable / None Entered.
Taxonomy
Not Applicable / None Entered.
Contact Email
