R Lee MacDonald1, Christopher G Thomas1,2,3,4, Alasdair Syme1,2,3. 1. Department of Physics and Atmospheric Science, Dalhousie University, Halifax, Nova Scotia, B3H 4R2, Canada. 2. Department of Medical Physics, Nova Scotia Health Authority, Queen Elizabeth II Health Sciences Centre, Halifax, Nova Scotia, B3H 1V7, Canada. 3. Department of Radiation Oncology, Dalhousie University, Halifax, Nova Scotia, B3H 4R2, Canada. 4. Department of Radiology, Dalhousie University, Halifax, Nova Scotia, B3H 4R2, Canada.
Abstract
PURPOSE: To develop an algorithm for dynamic collimator positioning to optimize beam's eye view (BEV) fitting of targets in dynamic conformal arc (DCA)-based radiotherapy procedures, of particular use in multiple metastases stereotactic radiosurgery procedures. METHODS: A trajectory algorithm was developed to dynamically modify the angle of the collimator as a function of arc-based control point to provide optimized collimation of target volume(s). Central to this algorithm is a concept denoted herein as "whitespace" defined as any nontarget area in the BEV that is not covered by any collimation system and is open to exposure from the radiation beam. Calculating whitespace at all collimator angles and every control point, a two-dimensional topographical map depicting the tightness-of-fit of the MLC was generated. A bidirectional gradient trajectory algorithm identified a number of candidate trajectories of continuous collimator motion. Minimization of integrated whitespace was used to identify an optimal solution for the navigation of the parameter space. Plans with dynamic collimator trajectories were designed for multiple metastases targets and were compared with fixed collimator angle dynamic conformal arc (DCA) plans and standard VMAT plans. RESULTS: Algorithm validation was performed on simple test cases with known solutions. The whitespace metric showed a strong correlation (R2 = 0.90) with mean dose to proximal normal tissue. Seventeen cases were studied using our algorithm to generate dynamic conformal arc (DCA) plans with optimized collimator trajectories for three and four target SRS patients and comparing them to DCA plans generated with optimized fixed collimator angles per arc and standard VMAT plans generated via template. Optimized collimator trajectories were found to produce a reduction in monitor units of up to 49.7 ± 5.1% when compared to VMAT across 17 patients, and all organ-at-risk and normal brain metrics were found to be superior or comparable. CONCLUSION: Dynamic collimator trajectories have the potential to improve DCA deliveries through increased efficiency, especially in the treatment of multiple cranial metastases. Implementation of this technology should not be hindered by mechanical safety considerations as collimator motions do not modify or introduce any new risks of collisions with patients.
PURPOSE: To develop an algorithm for dynamic collimator positioning to optimize beam's eye view (BEV) fitting of targets in dynamic conformal arc (DCA)-based radiotherapy procedures, of particular use in multiple metastases stereotactic radiosurgery procedures. METHODS: A trajectory algorithm was developed to dynamically modify the angle of the collimator as a function of arc-based control point to provide optimized collimation of target volume(s). Central to this algorithm is a concept denoted herein as "whitespace" defined as any nontarget area in the BEV that is not covered by any collimation system and is open to exposure from the radiation beam. Calculating whitespace at all collimator angles and every control point, a two-dimensional topographical map depicting the tightness-of-fit of the MLC was generated. A bidirectional gradient trajectory algorithm identified a number of candidate trajectories of continuous collimator motion. Minimization of integrated whitespace was used to identify an optimal solution for the navigation of the parameter space. Plans with dynamic collimator trajectories were designed for multiple metastases targets and were compared with fixed collimator angle dynamic conformal arc (DCA) plans and standard VMAT plans. RESULTS: Algorithm validation was performed on simple test cases with known solutions. The whitespace metric showed a strong correlation (R2 = 0.90) with mean dose to proximal normal tissue. Seventeen cases were studied using our algorithm to generate dynamic conformal arc (DCA) plans with optimized collimator trajectories for three and four target SRSpatients and comparing them to DCA plans generated with optimized fixed collimator angles per arc and standard VMAT plans generated via template. Optimized collimator trajectories were found to produce a reduction in monitor units of up to 49.7 ± 5.1% when compared to VMAT across 17 patients, and all organ-at-risk and normal brain metrics were found to be superior or comparable. CONCLUSION: Dynamic collimator trajectories have the potential to improve DCA deliveries through increased efficiency, especially in the treatment of multiple cranial metastases. Implementation of this technology should not be hindered by mechanical safety considerations as collimator motions do not modify or introduce any new risks of collisions with patients.