Room: AAPM ePoster Library
Purpose: TG-100 suggests that Quality Management (QM) programs utilize a risk-assessment based approach rather than a prescriptive approach. In light of TG-174’s focus on PET usage and QA in Radiation Therapy (RT) and our nine years of PET/CT RT QM experience, we investigated a framework for Failure Mode and Effects Analysis (FMEA) for PET/CT RT QM in the context of PET applications in target contouring.
Methods: Possible failure modes were identified by two medical physicists with input from a radiation oncologist. The severity (S) of a failure mode was assessed by its impact on contouring accuracy. The occurrence (O) and detectability (D) were derived from analysis of daily and monthly QA, retrospective review of clinical PET/CT images, physicists’ experience, expectation of staff performance, and possible causes of each failure mode. The Risk Priority Number (RPN) was calculated from O, S, and D using a 1 - 10 scale utilizing relative importance ranking. The first stage FMEA assumed no QM. A second stage FMEA re-evaluated the scores with our current quality control (QC)/QA, and the remaining high RPN failure modes were identified.
Results: Nine failure modes and eleven potential causes for these failure modes were specified. SUV offset (RPN=270), PET/CT misalignment (RPN=252), and excessive image noise (RPN=140) were the top three failure modes. Our current QC/QA procedures showed effectiveness in reducing risks of SUV offset (RPN=90), excessive image noise (RPN=80), and other image quality related failure modes (resolution, contrast, uniformity), but PET/CT misalignment (RPN=162) still remained with high RPN due to the fact that potential causes of this failure mode are currently not fully addressed by our QM.
Conclusion: FMEA was able to quantitatively assess current impact of our PET/CT RT QM. Future work includes implementation of a potential third stage with periodic re-evaluation FMEA using an institutional proposed QM program.