Development and Evaluation of Jaw Position Detection Method in Jaw-tracking Delivery Using an Electronic Portal...

Area of Science:

• Medical Physics
• Radiation Oncology
• Imaging Technology

Background:

• Accurate radiation delivery in radiotherapy is crucial for patient outcomes.
• Jaw-tracking delivery requires precise monitoring of jaw positions.
• Existing methods may lack the necessary accuracy or real-time feedback.

Purpose of the Study:

• To develop and validate a novel method for detecting jaw positions during jaw-tracking radiotherapy delivery.
• To ensure accurate radiation targeting using an electronic portal imaging device (EPID) in cine mode.
• To assess the accuracy of jaw position detection during dynamic treatment delivery.

Main Methods:

• Utilized a bidirectional picket fence (BPF) plan where jaws tracked multileaf collimator (MLC) apertures.
• Acquired irradiated patterns using an EPID in cine mode on a linear accelerator.
• Created a picket-fence-like image by summing cine images to evaluate MLC leaf and jaw positions.

Main Results:

• Jaw positions determined from cine EPID images showed high accuracy.
• Absolute differences for X1 and X2 jaws were 0.16 ± 0.19 mm and 0.11 ± 0.16 mm, respectively.
• Maximum error in detected MLC leaf positions was 0.11 mm, demonstrating sub-millimeter precision.

Conclusions:

• Jaw positions during jaw-tracking delivery can be accurately identified using cine EPID images.
• The developed method achieves accuracy better than 0.5 mm for jaw position detection.
• The BPF plan serves as a reliable picket fence test for verifying MLC leaf positions with high accuracy.