A portal dosimetry dose prediction method based on collapsed cone algorithm using the clinical beam model.

Area of Science:

• Medical Physics
• Radiation Oncology
• Imaging Technology

Background:

• Amorphous silicon electronic portal imaging devices (EPIDs) are crucial for dosimetric measurements in radiation therapy.
• Accurate dose prediction is essential for quality assurance and treatment verification.

Purpose of the Study:

• To determine if a portal dose prediction method, using the Pinnacle3 treatment planning system, can accurately calculate dose maps.
• To validate this method for amorphous silicon EPIDs in radiation therapy.

Main Methods:

• Developed a portal dose prediction method using CT images of a Varian aS1000 EPID as a quality assurance phantom within Pinnacle3.
• Calibrated the EPID, applied output factor corrections, and incorporated arm-backscattering corrections.
• Predicted and measured dose distributions for clinical treatments, comparing results with MatriXX measurements.

Main Results:

• The developed method successfully predicted dose maps using the existing linear accelerator model in Pinnacle3.
• CT imaging revealed a mean density of 1.16 g/cm³ for the EPID's sensitive area.
• Gamma index analysis (3%, 3 mm) showed high pass rates (≥97.4%) for both low and high modulated treatments, comparable to MatriXX results.

Conclusions:

• The portal dose prediction method is easily implementable in clinics without requiring new energy modeling or independent algorithms.
• A key advantage is using the same algorithm and beam model for both patient dose calculation and portal dose prediction.
• Independent validation with an ionization chamber matrix confirmed the method's accuracy.