Using a Novel Leaf Sequencer and Segment Shape Optimization Algorithm to Reduce Treatment Session Times On the Elekta Unity
J Snyder*, J St-Aubin, S Yaddanapudi, S Strand, S Kruger, R Flynn, D Hyer, University of Iowa Hospitals and Clinics, Iowa City, IA
Presentations
PO-GePV-T-206 (Sunday, 7/25/2021) [Eastern Time (GMT-4)]
Purpose: A current limitation of MR-guided radiotherapy is extended treatment session times. This study evaluates a new Elekta developed research optimizer and leaf sequencing algorithm with respect to plan quality and the potential to reduce optimization as well as treatment delivery times for Elekta Unity patients.
Methods: Three prostate, three oligometastases, two pancreatic, and two liver patients previously treated on the MR-linac were retrospectively evaluated. Reference plans for all ten patients were generated using the clinical Hyperion optimizer and leaf sequencer. These plans were then re-optimized using a novel Optimal Fluence Levels (OFL) leaf sequencing algorithm and Pseudo Gradient Descent optimization algorithm (OFL+PGD). Beam angles and IMRT objectives were held constant for clinical and research algorithm optimizations. All plans were normalized for equivalent PTV coverage. Optimization time, treatment delivery time, number of plan segments, total MU, and gamma passing rate were all evaluated. A paired t-test was used to determine statistical significance. Relative changes in plan quality were evaluated for the PTV and critical OAR’s.
Results: The PGD and OFL algorithms had a reduction in optimization time of 51.4 ± 5.0% and a reduction in treatment time of 10.6 ± 7.5% as compared to the Hyperion plans. The OFL+PGD plans used 5.6 ± 9.1% fewer MU, 13.0 ± 12.4% fewer plan segments, and had an average gamma passing rate of 99.1% compared to 99.2% for the Hyperion algorithm. Reductions in optimization time (p=0.002), delivery time (p=0.005), and number of plan segments (p=0.009) are statistically significant. The plan quality between the Hyperion and OFL+PGD algorithms was observed to be equivalent in terms of OAR sparing and PTV hotspot.
Conclusion: The OFL+PGD generated plans optimized faster and had shorter delivery times while maintaining equivalent plan quality. These algorithms have the potential to decrease overall treatment session times for MR-guided radiotherapy.
Funding Support, Disclosures, and Conflict of Interest: Funding for this work was supplied through an industry grant from Elekta
ePosters
Keywords
Treatment Planning, Optimization
Taxonomy
IM/TH- MRI in Radiation Therapy: MRI/Linear accelerator combined dose optimization
Contact Email