(Sunday, 7/12/2020) [Eastern Time (GMT-4)]
Room: AAPM ePoster Library
Purpose: To quantify the improvement of dose conformity and targeting accuracy with the use of control point-specific image registration when treating very small lesions. This is evaluated using virtual cones with simulated intrafraction patient motion in a phantom.
Methods: A single-sheet of Gafchromic film was inserted in the MAX-HD™ phantom with a metal-bb attached to the film. Treatment consisted of two coplanar arcs with a 2.1 mm x 5 mm field shaped by a multi-leaf collimator. Two arcs were delivered with the collimator angle at ± 45°. 50 monitor units were delivered every 10 degrees of gantry movement. The phantom was linearly moved away from its initial position during treatment until it was 1.44 mm offset along each axis by the end of treatment (2.5 mm 3D offset). MV images were acquired at each control point and mutual information was used to re-align the phantoms’ anatomy to isocenter with couch shifts. Films were scanned at 400 dpi and converted to dose distributions using triple channel film dosimetry.
Results: The maximum dose was increased by 8.6% when anatomical image registration and control point-specific positional corrections were applied compared with no phantom realignment. The dose area was ~19% smaller at the 90% isodose, ~ 13% at the 70% isodose, and ~9% smaller at the 50% isodose when compared with no phantom realignment. Targeting accuracy in a 2D-plane (i.e. planned vs delivered dose centre of mass) was found to be 1.55 mm, and 0.18 mm for treatment without phantom realignment, and anatomical registration respectively.
Conclusion: This work demonstrates that control point-specific image registration has the potential to improve treatment outcomes by minimizing dosing-errors during due to intrafraction patient motion. Future work will investigate the dosimetric impacts of different types of intrafraction motion such as large shifts that occur between control points.
Keywords
Not Applicable / None Entered.
Taxonomy
Not Applicable / None Entered.
Contact Email
