Gamma Knife (GK) treatments are typically delivered in 1 day with little opportunity to test different planning strategies. The authors demonstrate 2 methods for imposing GK coordinate systems upon imaging datasets without frame attachment to allow leisurely preprocedural planning, and discuss potential applications.
A “virtual framing” is constructed by coregistering a CT scan of a Leksell frame with a patient dataset using the GammaPlan (Multiview) module. Equations for skull radii are derived by approximating the skull as an ellipsoid. No proprietary software other than that of the GK system is required. In a second method, images of fiducial markers are directly superimposed on the patient dataset. Validation of the first method was achieved by comparing the lengths of 75 line segments and 60 single shot diameters measured in the virtual coordinates with those measured in real coordinates. In addition to preplanning, 2 applications are discussed. The first is the use of GK software to aid radiosurgical planning for other devices. The second is the use of virtual framing to enhance automatic optimization algorithms.
Mean (± standard deviation) and root-mean-square differences in lengths were 0.18 ± 0.32 and 0.37 mm. Mean and root-mean-square differences in diameters of single-shot plans were 0.01 ± 0.18 and 0.18 mm.
Virtual framing allows exploration of radiosurgical planning strategies prior to the day of treatment using only the GK software. Other applications include enhancement of radiosurgical planning for other systems and enhancement of optimization algorithms.