Assessing the validity of simulation-based sensitivity analysis for patient-specific quality assurance devices

Area of Science:

• Medical Physics
• Radiotherapy
• Quality Assurance

Background:

• Patient-Specific Quality Assurance (PSQA) is vital for radiotherapy accuracy and safety.
• Assessing PSQA system sensitivity to errors lacks standardized methods.
• Current methods include Direct Error (DE) and Inverse Error (IE) approaches.

Purpose of the Study:

• Compare DE and IE methods for EPID-based PSQA sensitivity analysis.
• Evaluate if the IE method can replace the more time-consuming DE method.
• Determine reliability for error detection and classification in radiotherapy.

Main Methods:

• Utilized nine treatment plans (3DCRT, VMAT) on a thoracic phantom.
• Introduced delivery inaccuracies, setup misalignments, and anatomical variations.
• Evaluated dosimetric differences and analyzed sensitivity/specificity.

Main Results:

• DE and IE methods showed good agreement for delivery inaccuracies but varied for patient errors.
• Higher thresholds (≥10%) improved agreement for anatomical/delivery errors, but not setup errors.
• The IE method overestimated sensitivity at lower thresholds, especially for positioning errors.

Conclusions:

• The IE method is a viable alternative for PSQA sensitivity analysis.
• IE and DE methods are not fully interchangeable for all error types.
• Recommend validating the IE method before widespread adoption in radiotherapy.