SU-E-CAMPUS-T-04: Statistical Process Control for Patient-Specific QA in Proton Beams

Authors


Abstract

Purpose:

To evaluate and improve the reliability of proton QA process, to provide an optimal customized level using the statistical process control (SPC) methodology. The aim is then to suggest the suitable guidelines for patient-specific QA process.

Methods:

We investigated the constancy of the dose output and range to see whether it was within the tolerance level of daily QA process. This study analyzed the difference between the measured and calculated ranges along the central axis to suggest the suitable guidelines for patient-specific QA in proton beam by using process capability indices. In this study, patient QA plans were classified into 6 treatment sites: head & neck (41 cases), spinal cord (29 cases), lung (28 cases), liver (30 cases), pancreas (26 cases), and prostate (24 cases).

Results:

The deviations for the dose output and range of daily QA process were ±0.84% and ±019%, respectively. Our results show that the patient-specific range measurements are capable at a specification limit of ±2% in all treatment sites except spinal cord cases. In spinal cord cases, comparison of process capability indices (Cp, Cpm, Cpk ≥1, but Cpmk ≤1) indicated that the process is capable, but not centered, the process mean deviates from its target value. The UCL (upper control limit), CL (center line) and LCL (lower control limit) for spinal cord cases were 1.37%, −0.27% and −1.89%, respectively. On the other hands, the range differences in prostate cases were good agreement between calculated and measured values. The UCL, CL and LCL for prostate cases were 0.57%, −0.11% and −0.78%, respectively.

Conclusion:

SPC methodology has potential as a useful tool to customize an optimal tolerance levels and to suggest the suitable guidelines for patient-specific QA in clinical proton beam.

Ancillary