Error detection using EPID-based 3D in vivo dose verification for lung stereotactic body radiotherapy

Area of Science:

• Medical Physics
• Radiation Oncology

Background:

• Stereotactic body radiation therapy (SBRT) for lung cancer requires precise dose delivery.
• In vivo dosimetry verification systems are crucial for ensuring treatment accuracy.
• Epidermal electronic portal imaging devices (EPIDs) offer a potential solution for real-time dosimetry.

Purpose of the Study:

• To evaluate the error detection capabilities of an EPID-based 3D in vivo dosimetry system.
• To identify the limitations of this system in detecting various errors during lung SBRT.
• To assess the system's suitability for clinical implementation in lung SBRT.

Main Methods:

• Intentionally introduced 30 dynamic and constant machine errors into lung SBRT plans.
• Delivered error plans to a CIRS phantom using SBRT technique.
• Reconstructed dose distributions using iViewDose™ and compared with error-free plans.

Main Results:

• The system successfully detected errors in central leaf positions, dynamic MLC, jaw movements, setup shifts, and anatomical changes.
• Dynamic gantry angle and collimator angle errors were not detected due to the rotation-symmetric target shape.
• Gamma analysis indicators (γmean and γpassrate) detected 81.3% and 87.5% of errors, respectively (excluding gantry, collimator, and output errors).

Conclusions:

• The iViewDose™ system is suitable for detecting most clinical errors in lung SBRT.
• Detectability limitations exist for dynamic gantry angle and constant collimator angle errors.
• Further system refinement may be needed to address specific error detection gaps.