Encrypted login | home

Program Information

Asymmetries in the CyberKnife Iris 2 Collimator


G Cernica

G Cernica*, H Ji , D McRae , Associates In Medical Physics, Lanham, MD and Virginia Hospital Center, Arlington VA

Presentations

SU-F-T-588 (Sunday, July 31, 2016) 3:00 PM - 6:00 PM Room: Exhibit Hall


Purpose: The Iris 2 dodecahedral collimator in the CyberKnife unit can demonstrate visible asymmetry in the shaped radiation field, as seen on routine films. Specifically some edges of the collimator project longer than others. PDDs and output factors remain unaffected by this asymmetry. An in-house program was written to analyze the impact of this asymmetry on profiles.

Methods: Two years of routine chromodynamic films were retrospectively analyzed using an in-house developed program. Films were obtained on a weekly basis in a plastic phantom on the vendor-provided mount. Profiles were obtained every degree. A dodecahedron was fitted to the 50% isodose line and the positions and sizes of the upper and lower hexagonal collimators were derived. The collimators’ profiles at every degree, symmetry and centering were calculated and trended over the months.

Results: Asymmetries in the obtained profiles were apparent. The upper and lower collimators were off-centered (0.31 mm, p<10-8), they did not project the same field size at 80 cm (0.46 mm difference, p<10-31). Profiles as a function of angle demonstrate a periodicity, but short and long profiles are rarely 15° apart, but vary from 9° to 21°. In addition, a second, intermediate minimum is seen at every other short profile. A trend in the field size over time is noted and can be monitored to identify servicing needs.

Conclusion: Despite its differences, the asymmetric Iris 2 collimator did not demonstrate large deviations in PDDs and output factors. Due to the asymmetry, the 15° apart profiles obtained at commissioning do not represent the true large and short blade profiles of the collimator. Averaging of the profiles in the commissioning process hides this difference.



Contact Email: