Evaluation and comparison of second-check monitor unit calculation software with Pinnacle³ treatment planning system

The purpose of this study was to evaluate and compare the accuracy of dose calculations in second check softwares (Diamond, IMSure, MuCheck, and RadCalc) against the Phillips Pinnacle³ treatment planning system. Eighteen previously treated patients' treatment planning files consisting of a total of 204 beams were exported from the Pinnacle³ TPS to each of the four second check software. Of these beams, 145 of the beams used were IMRT plans while 59 were VMAT arcs. The values were represented as a percent difference between primary and secondary calculations and used for statistical analysis. Box plots, Pearson Correlation, and Bland-Altman analysis were performed in MedCalc. The mean percent difference in calculated dose for Diamond, IMSure, MuCheck, and RadCalc from Pinnacle³ were -0.67%, 0.31%, 1.51% and -0.36%, respectively. The corresponding variances were calculated to be 0.07%, 0.13%, 0.08%, and 0.03%; and the largest percent differences were -7.9%, 9.70%, 9.39%, and 5.45%. The dose differences of each of the second check software in this study can vary considerably and VMAT plans have larger differences than IMRT. Among the four second check softwares, RadCalc values has shown a high agreement on average with low variation, and had the smallest percent range from Pinnacle³ values. The closest in average percent difference from the Pinnacle³data was the IMSure software, but suffered from significantly larger variance and percent range. The values reported by Diamond and MuCheck had significantly high percent differences with TPS values.