SU‐F‐P‐07: Applying Failure Modes and Effects Analysis to Treatment Planning System QA

Abstract

Purpose:A small‐scale implementation of Failure Modes and Effects Analysis (FMEA) for treatment planning system QA by utilizing methodology of AAPM TG‐100 report.Methods:FMEA requires numerical values for severity (S), occurrence (O) and detectability (D) of each mode of failure. The product of these three values gives a risk priority number (RPN). We have implemented FMEA for the treatment planning system (TPS) QA for two clinics which use Pinnacle and Eclipse TPS. Quantitative monthly QA data dating back to 4 years for Pinnacle and 1 year for Eclipse have been used to determine values for severity (deviations from predetermined doses at points or volumes), and occurrence of such deviations. The TPS QA protocol includes a phantom containing solid water and lung‐ and bone‐equivalent heterogeneities. Photon and electron plans have been evaluated in both systems. The dose values at multiple distinct points of interest (POI) within the solid water, lung, and bone‐equivalent slabs, as well as mean doses to several volumes of interest (VOI), have been re‐calculated monthly using the available algorithms.Results:The computed doses vary slightly month‐over‐month. There have been more significant deviations following software upgrades, especially if the upgrade involved re‐modeling of the beams. TG‐100 guidance and the data presented here suggest an occurrence (O) of 2 depending on the frequency of re‐commissioning the beams, severity (S) of 3, and detectability (D) of 2, giving an RPN of 12.Conclusion:Computerized treatment planning systems could pose a risk due to dosimetric errors and suboptimal treatment plans. The FMEA analysis presented here suggests that TPS QA should immediately follow software upgrades, but does not need to be performed every month.