Recommended Citation
Mann TD, Thind KS, and Ploquin NP. Fast stereotactic radiosurgery planning using patient-specific beam angle optimization and automation. Phys Imaging Radiat Oncol 2022; 21:90-95.
Document Type
Article
Publication Date
1-1-2022
Publication Title
Phys Imaging Radiat Oncol
Abstract
BACKGROUND AND PURPOSE: Linac-based stereotactic radiosurgery (SRS) planning for multi-metastatic cases is a complex and intensive process. A manual planning strategy starts with a template-based set of beam angles and applies modifications though a trial and error process. Beam angle optimization uses patient specific geometric heuristics to determine beam angles that provide optimal target coverage and avoid treating through Organs-at-Risk (OARs). This study expands on a collision prediction application developed using an application programming interface, integrating beam angle optimization and collision prediction into a Stereotactic Optimized Automated Radiotherapy (SOAR) planning algorithm.
MATERIALS AND METHODS: Twenty-five patient plans, previously treated with SRS for multi-metastatic intracranial tumors, were selected for a retrospective plan study comparing the manual planning strategy to SOAR. The SOAR algorithm was used to select isocenters, table, collimator, and gantry angles, and target groupings for the optimized plans. Dose-volume metrics for relevant OARs and PTVs were compared using double-sided Wilcoxon signed rank tests (α = 0.05). A subset of five patients were included in an efficiency study comparing manual planning times to SOAR automated times.
RESULTS: OAR dose metrics compared between planning strategies showed no statistical difference for the dataset of twenty-five plans. Differences in maximum PTV dose and the conformity index were improved for SOAR planning and statistically significant. The median SOAR planning time was 9.8 min compared to 55 min for the manual planning strategy.
CONCLUSIONS: SOAR planning was comparable in plan quality to a manual planning strategy with the possibility for greatly improving planning efficiency through automation.
PubMed ID
35243038
Volume
21
First Page
90
Last Page
95
