SU-E-T-16: Statistical Variability and Confidence Intervals for Planar Dose QA Pass Rates.

Area of Science:

• Medical Physics
• Radiation Oncology
• Dosimetry

Background:

• Comparing measured and calculated dose planes commonly uses pass rates based on percent difference and distance-to-agreement (DTA).
• Low areal-density detector arrays are often used, leading to statistical uncertainty in pass rate calculations.

Purpose of the Study:

• To examine the statistical uncertainty of planar dose comparison pass rates.
• To propose methods for establishing confidence intervals for pass rates obtained with low detector-density arrays.

Main Methods:

• Acquired absolute dose planes using EPID for intensity-modulated fields.
• Simulated low-density grids by sampling high-density EPID data with varying detector densities and orientations.
• Calculated pass rates using various composite analysis techniques for each simulation.

Main Results:

• Pass rates exhibit a distribution of possible values for each dose plane pair, regardless of detector grid uniformity.
• Binomial distribution can predict these distributions and approximate confidence intervals based on sampling density and observed pass rate.
• Example: 95% confidence intervals for IMRT pass rates (2%,2mm) average +/-5.3% and +/-3.8% with 1 and 2 detectors/cm² grids, respectively.

Conclusions:

• Pass rates from low-density array measurements are not absolute and require reporting with calculation methods and confidence intervals.
• These findings are applicable to 3D detector arrays.
• Fixed 'action levels' for pass rates need reevaluation for low-density array measurements.