A statistical approach to IMRT patient-specific QA

Area of Science:

• Medical Physics
• Radiation Oncology
• Statistical Quality Control

Background:

• Intensity modulated radiation therapy (IMRT) patient-specific quality assurance (QA) is resource-intensive.
• A single global QA tolerance may lead to excessive false negatives or positives across diverse treatment sites.
• This can result in wasted resources, treatment delays, and compromised efficiency.

Purpose of the Study:

• To develop a tool for identifying statistical variations, monitoring trends, and detecting outliers in IMRT QA.
• To analyze IMRT QA data to understand site-specific differences and evaluate the need for tailored tolerance levels.
• To reduce false negatives and false positives in QA measurements.

Main Methods:

• Analysis of QA measurements using two ion chamber points.
• Development of a custom software tool for data processing, retrieval, visualization, and statistical analysis.
• Application of analysis of variance (ANOVA) and statistical process control (SPC) techniques.

Main Results:

• Retrospective analysis confirmed significant differences between treatment sites, supporting the need for site-specific QA tolerances.
• The developed tool effectively monitored QA process variability using I, S, and exponentially weighted moving average (EWMA) charts.
• Site-specific tolerances were identified as crucial for reducing false positive and false negative QA results.

Conclusions:

• Analysis of variance on ion chamber measurements supports the need for site-specific QA tolerances in IMRT.
• A method for calculating appropriate site-specific tolerances was proposed and illustrated for clinical use.
• The developed automated tool aids in reducing process variability and uncertainty during IMRT commissioning.