Evaluating the sensitivity of Halcyon's automatic transit image acquisition for treatment error detection: A phantom...

Area of Science:

• Medical Physics
• Radiotherapy Technology
• Image Analysis

Background:

• The Varian Halcyon™ system features an in-line electronic portal imaging detector that captures transit images for all patients.
• These images offer potential for "every patient, every monitor unit" quality assurance and adaptive radiotherapy.
• Evaluating the imager's sensitivity to clinical errors and daily variations is crucial for its clinical application.

Purpose of the Study:

• To assess the Varian Halcyon™ electronic portal imaging detector's sensitivity to potential clinical errors.
• To evaluate the imager's response to day-to-day variations using clinical exit images.
• To establish the feasibility of using transit images for quality assurance and adaptive radiotherapy.

Main Methods:

• Simulated clinical errors including output variations (2%-10%), buildup changes (0.5-5.0 cm), and phantom shifts (2-10 cm lateral, 0.2-1.5 cm other).
• Analysis of mean relative differences (MRDs) and standard deviations in pixel-difference histograms (σRD) between test and baseline images.
• Evaluation of daily exit images from six prostate patients to assess variations with and without gas presence.

Main Results:

• MRDs showed linear response to output and buildup changes, detecting 1% output change and 0.2 cm buildup change with 2.5σ confidence.
• Lateral shifts were accurately calculated (within 0.5 mm) for heterogeneous phantoms using cross-correlation.
• Significant associations between MRD/σRD and gas presence were observed in five of six patients.

Conclusions:

• Automated analysis of Halcyon™ exit images demonstrates high sensitivity for detecting mid-treatment changes.
• Establishing appropriate detection thresholds is necessary for clinical implementation.
• This study provides foundational steps for developing automated image evaluation in radiotherapy.