A framework for automated and streamlined kV cone beam computed tomography image quality assurance: a...

Area of Science:

• Medical Physics
• Radiology
• Quality Assurance

Background:

• Cone Beam Computed Tomography (CBCT) is crucial for modern radiotherapy.
• Variability in CBCT quality assurance (QA) practices across institutions can impact clinical outcomes.
• Standardized and automated QA is needed to ensure consistent image quality.

Purpose of the Study:

• To develop and evaluate a framework for automated, standardized CBCT image quality QA.
• To assess the variability of CBCT QA practices among institutions.
• To compare CBCT image quality metrics across multiple linear accelerators (linacs).

Main Methods:

• A survey was conducted to understand existing CBCT QA practices.
• A commercial, automated software platform was validated by seven institutions.
• A MATLAB-based platform automated image upload and analysis.
• CBCT image quality metrics were analyzed using control charts.

Main Results:

• The automated framework successfully standardized CBCT QA across seven institutions and 23 linacs.
• Institutions often employed stricter QA goals for uniformity, contrast, and HU constancy than recommended.
• Quantitative, longitudinal assessment of image quality metrics was achieved.

Conclusions:

• The developed framework supports automated, standardized CBCT image quality QA.
• The study identified that institutional QA goals are frequently more stringent than established guidelines.
• This work provides a reference for machine-specific tolerance optimization and risk-based QA programs.