Exhibit Hall | Forum 7
Purpose: The objective of this study is to characterize the Fast Inverse Planning optimizer in Gamma Knife (GK) system and determine the optimization parameters that result in the optimal stereotactic radiosurgery (SRS) plan quality for benign tumors and arteriovenous malformations(AVM).
Methods: Twenty-three patients treated with GK SRS for Vestibular Schwannoma (VS- 9 cases), AVM(8 cases), and Pituitary Adenoma (PA- 6 cases), were evaluated. Each target was planned by varying the low dose (LD) and beam-on-time (BOT) penalties in 0.1 increments from 0 to 1. Single fraction prescription doses were 13 Gy, 20 Gy, and 24 Gy for VS, AVM and PA lesions, respectively. All plans were optimized to achieve a minimum target coverage of 99%. The following plan quality metrics were recorded for each plan: Paddick conformity index (PCI), gradient index (GI), BOT and maximum organ-at-risk (OAR) doses. An objective score matrix was calculated for each plan using a linear weighted combination of the aforementioned metrics. A histogram of optimal solutions containing the 10 best score values was extracted.
Results: : 121 plans per patient generated for the range of LD and BOT penalties totaling 2783 plans were included in this analysis. The range of PCI, BOT, and GI across all targets were 0.13 to 0.94, 7.6 to 540 minutes, and 2.2 to 5.84, respectively. Optimal solutions for the benign lesions were found to occupy the solution space just below the diagonal, and specifically, the histogram showed that highest frequencies occurred at the LD/BOT penalty settings of (0.2, 0.4), (0, 0.3) and (0.1, 0.4).
Conclusion: In this study, an analysis of the user-inputted variables were systematically varied to determine the optimal solutions to maximize PCI and minimize GI, OAR doses and BOT simultaneously and provide guidance for SRS planners to generate plans for benign lesions more efficiently.