Session Item

SRS and HyperArc stereotactic radiosurgery techniques deliver multiple VMAT arcs with fixed couch angles. Dosimetric verification of this technique is time consuming, and it is difficult to verify multiple target deliveries adequately. A new method was developed where EPID frames captured during each arc are used to reconstruct 3D dose in a 20 cm diameter water equivalent virtual spherical phantom (VSP) incorporating couch rotations.

The VSP geometry is invariant under couch and gantry rotation which simplifies dose reconstruction. The SRS/hyperarc plan is transferred to the VSP in the treatment planning system (TPS) with couch angles and dose calculated (Eclipse V15.6.05/15.6.8). EPID image frames are captured on Varian Truebeam linear accelerators in-air (without couch rotations) with aS1200 EPID and used to construct cine EPID images after insertion of the arc couch angle in the image header. For each acquired image, 3D dose is calculated in the VSP at the corresponding gantry and couch angle using a depth-dependent EPID to dose conversion model and all image doses summed. TPS and reconstructed dose are compared using 3D gamma analysis. To experimentally validate the method, EPID reconstructed dose was compared to measured gafchromic film measurement in the Lucy SRS phantom (14 cm diameter) for a 5-lesion (each 10 mm diameter) HyperArc test plan after recalibrating the method to estimate dose in the Lucy phantom. Clinical results for EPID reconstructed dose distributions compared to TPS  were obtained for 36 patients (52 plans) for 6FFF energy using 3D gamma analysis in the VSP.

The comparison of film and EPID reconstructed dose in the Lucy phantom is shown in Figure 1, along with profile comparison to the TPS dose. A 2D gamma pass-rate of 97.1% was obtained with 5%, 1 mm criteria and 10% dose threshold. The pass-rate for measured compared to TPS dose was 99.9%. The EPID method calculates for a perfect uniform density phantom while the Lucy phantom has significant shape and density variations resulting in greater differences than for the TPS, however the results confirm that the method is reproducing measured dose. An example of a clinical verification result is shown in Figure 2. The comparisons between EPID reconstruction and TPS gave gamma pass-rates (mean ± 1 SD) of 99.3% ±0.59% and a minimum of 97.7% at 3%, 1 mm, 10% dose threshold criteria.

SRS and Hyperarc are challenging to validate with measurement due to the multiple lesion targets. A new method has been developed for verification of SRS/Hyperarc that is efficient to perform and enables rigorous 3D assessment of the delivered dose distribution. It has been shown to accurately reproduce measured dose in phantom and the the 3D dose distributions in the VSP show a high level of agreement to TPS dose distributions.