MLC quality assurance using EPID: a fitting technique with subpixel precision.

Area of Science:

• Medical Physics
• Radiotherapy Technology
• Image Analysis

Background:

• Electronic portal imaging devices (EPIDs) offer an alternative to film for quality assurance (QA) of multileaf collimator (MLC) positioning.
• Accurate MLC positioning is critical for effective radiation therapy and patient safety.

Purpose of the Study:

• To present a novel method for acquiring and analyzing EPID images to precisely determine MLC leaf positioning accuracy.
• To validate the accuracy and robustness of the proposed EPID-based QA method.

Main Methods:

• Utilized amorphous silicon EPIDs to capture strip-test images of MLCs.
• Developed analytical fits for interleaf leakage and leaf abutment signatures to extract positional data.
• Employed multiobjective optimization and modified Lorentzian functions for image rotation, leaf gap width, and position parameterization.

Main Results:

• Achieved high accuracy in determining collimator rotation angle (0.01 degrees) and MLC leaf gap width (0.2 mm changes detected).
• Demonstrated subpixel resolution for MLC leaf position variations, with a precision of 0.1 mm at a 95% confidence level.
• The method proved robust against image noise and pixel size variations.

Conclusions:

• The presented EPID-based method provides a reliable and accurate approach for routine MLC QA.
• This technique can assist physicists in establishing effective action levels for MLC quality assurance.
• The findings contribute to improving the precision and safety of radiation therapy delivery.