Program Information
Planning Margin Implications for Multiple-Target, Single-Isocenter VMAT SRS Based On Image Guidance Tolerances
D Harrington1*, E Clouser1 , (1) Mayo Clinic Arizona, Phoenix, AZ
Presentations
SU-D-BRB-3 (Sunday, July 12, 2015) 2:05 PM - 3:00 PM Room: Ballroom B
Purpose: Single-isocenter, multiple target VMAT SRS for multiple brain metastases typically features field isocenter positional displacement from the individual lesion centroids. This study examines the planning margin implications of this isocenter displacement for VMAT SRS owing strictly to verification tolerances of the image guidance system.
Methods: SRS image guidance at this institution is provided by Brainlab ExacTrac in combination with 6DOF couch. The system provides patient position correction through a combination of couch rotations and translations. It is well-known that affine transformations can be compactly represented in matrix form using homogeneous coordinates. The maximum geometric error directly attributable to the x-ray verification tolerances specified in ExacTrac was determined in homogenous coordinates for a range of practical radial distances from isocenter to lesion centroid. These tolerances are selectable and are based on typical uncertainties achieved by the method of patient immobilization. The calculated maximum geometric error is dependent on the understanding that the translational and rotational tolerances selected for x-ray verification represent the ability of the image guidance system to correct errors up to, and including, the specified tolerance. The additional planning margin in the worst-case is assumed to be the magnitude of the maximum geometric error.
Results: The maximum error is a non-linear function of the radial distance from field isocenter to lesion centroid. It is dependent on the angular and translational tolerances specified based on setup uncertainties associated with typical immobilization options for SRS. The typical institutional SRS tolerance of 0.5 mm and 0.5° consistent with Brainlab mask immobilization yields maximum error of 0.867 – 1.23 mm over the range 0 – 7 cm radial distance.
Conclusion: An expression for the maximum error attributable to the image guidance system for multiple-target, single-isocenter VMAT SRS was developed, enabling proper specification of planning margins to account for the associated geometric uncertainty.
Contact Email: