Quality assurance of a gimbaled head swing verification using feature point tracking

Area of Science:

• Medical Physics
• Radiotherapy Technology
• Image Guidance

Background:

• Dynamic tumor tracking (DTT) requires precise and safe gimbaled head swing verification for clinical accuracy.
• Daily quality assurance (QA) is essential to maintain the performance of DTT systems.

Purpose of the Study:

• To develop and validate a quantitative method for daily QA of gimbaled head swing using readily available technology.
• To ensure the reliability and safety of DTT systems for clinical applications.

Main Methods:

• A web camera was positioned on the couch to record gimbaled head motion (sinusoidal pattern, ±20 mm amplitude, 3 s cycle) in pan and tilt directions.
• Feature point tracking, specifically the pyramidal Lucas-Kanade optical flow method, was applied to a tapped hole on the gimbaled head.
• Analysis of image data quantified the period and amplitude of the gimbaled head swing for daily QA.

Main Results:

• The mean period of the gimbaled head swing was consistently around 3.00 s (±0.03 s pan, ±0.02 s tilt).
• The mean relative displacement closely matched the target amplitude, with 19.7 ± 0.08 mm in the pan and 18.9 ± 0.2 mm in the tilt direction.
• The results demonstrated the reliability of the gimbaled head swing for DTT applications.

Conclusions:

• A quantitative, webcam-based method for daily QA of gimbaled head swing has been successfully developed.
• This method provides reliable assessment of gimbaled head swing against baseline values for DTT systems.
• The proposed technique enhances the safety and accuracy of clinical DTT through robust daily quality assurance.