Feasibility of Using Machine Performance Check for Daily Photon Output Constancy, and Isocentricity QA

Document Type

Conference Proceeding

Publication Date


Publication Title

Med Phys


Purpose: Machine Performance Check (MPC) utilizes the electronic portal imaging device (EPID) to perform geometric and beam checks, including monitoring of linac output changes, as well as distance between KV imaging and radiation isocenters (DKV-MV). Here we investigated the accuracy of MPC for evaluating daily output constancy. The accuracy of DKV-MV reported by MPC versus Winston-Lutz measurements, was also tested. Methods: MPC was compared with daily output measurements using a Sun Nuclear DailyQA3 device for 6X, 10X, 15X and 10XFFF over 7 months (N = 138), as well as monthly output measurements using an ADCL-calibrated farmer-type ion chamber in water. MPC checks were redone after EPID recalibration (dark/flood fields, all imaging modes, dosimetry normalization). DKV-MV were compared against Winston-Lutz measurements of isocentricity, acquired at four cardinal gantry angles over 97 days. Results: Absolute percent differences between MPC and DailyQA3 (mean±SD) were 0.26±0.37, 0.12±0.44, 0.33±0.42 and 0.14 ± 0.43 for 6X, 10X, 15X and 10XFFF, respectively. All three methods detected output drifts, and trends were similar amongst different energies. MPC was in good agreement with variations in monthly output ion chamber measurements, while DailyQA3 over-responded to output changes. The linear regression coefficients of the increase in output (%/day) over 7 months were 0.0040 ± 0.0005, 0.0045 ± 0.0001 and 0.0072 ± 0.0004 for ion chamber, MPC and DailyQA3, respectively. MPC outputs were -0.28% ± 0.44 and -0.33%±0.23 before and after calibration, respectively. For isocentricity, the average difference between DKV-MV (MPC) and Winston-Lutz was 0.02±0.10 mm. Conclusion: True variations in machine output (as measured monthly with an ion chamber and daily with DailyQA3) were accurately captured by MPC. Re-calibration of the EPID did not impact on MPC results. For the purpose of detecting congruence between KV imaging and radiation isocenters, MPC was in excellent agreement with Winston-Lutz measurements. MPC appears to be an accurate tool for monitoring of output consistency and linac/imager isocentricity.





First Page


Last Page


This document is currently not available here.