68-4

61/136

,V95%, D95%, and Dmax of the semiautomatic FIF and original plans and those of the semiautomatic FIF and manual FIF plans were statistically compared. Since these data did not follow a normal distribu-tion, the Wilcoxon signed-rank test was performed using the JMP software (version 12.2.0; SAS Insti-tute Inc., USA). A P-value of <0.05 was accepted as significant.D98%follows:by normalizing the difference between D2% and D98% with D50%Figure 6　The box plots for (a)V105%, (b)V95%, (c)D95%, (d)Dmax, and (e) HI in the original, manual field-in-field (FIF), and semiautomatic FIF plans. The box plot for the processing time in the manual FIF and semiautomatic FIF plans is also presented in (f). p indicates P-values for the Wilcoxon signed-rank test.16).Measurement of the processing time for creating the sub-beams and adjusting the weightsThe manual FIF creation time was measured with a stopwatch from the start of FIF creation to the end of the dose calculation. The elapsed time between the opening of the original plan in the TPS and the final dose calculation was measured in the automatic FIF script. The processing time included the time for creating sub-beams, adjusting the beam weights, and dose calculation. Comparison of dose indices and processing timeA total of 21 cases resulted in FIF and one case in FIF-4SF. Figure 6a-e shows the box plots of V105%, V95%, D95%, Dmax, and HI in the original, manual FIF, and semiautomatic FIF plans. The creation 381=HID2%D50%Results