Evaluation of the Performance of the Segmentation Methods Used in the CMS XiO Step-And-Shoot IMRT
I Jurkovic1*, P Mavroidis2,3, S Stathakis2, C Esquivel2, V Chyle4, N Papanikolaou2, (1)University of Texas Health Science Center, UTHSCSA, San Antonio, TX, (2) Cancer Therapy and Research Center University of Texas Health Sciences Center, San Antonio, TX, (3) Karolinska Institutet and Stockholm University, Sweden, (4) Kerrville Cancer Center, Kerrville, TXSU-E-T-610 Sunday 3:00PM - 6:00PM Room: Exhibit Hall
Purpose: The aim of this work is to evaluate the performance of the segmentation methods used in the CMS XiO radiotherapy treatment planning system.
Methods:For this study, fifteen patients with different types of cancer (prostate, head and neck, pancreas, esophagus, lung) were selected. For each patient, two treatment plans were produced using the sliding window segmentation and the smart sequencing methods. The plans were optimized based on the same clinical objectives. The analysis was based on the dose volume histograms (DVHs) of the PTV and organs at risk, the isodose distributions, maximum global dose, total number of segments, total number of monitor units (MU) per plan, and the average number of MUs per segment.
Results:The study showed a statistically significant reduction in the total number of monitor units (mean:25.1%, range:7.0%-35.5%) when the smart sequencing segmentation method was used (p=1.2E-7). However, this reduction in the total number of monitor units did not reduce the total number of MLC segments per plan. The number of MLC segments per plan varied regardless of the site being treated. The plans that used the smart sequencing segmentation method had higher global maximum doses ranging between 106%-118% of the prescription dose, while the corresponding range for the sliding window method was 104%-111%. The average number of monitor units per MLC segment was 2.5 for the smart sequencing method (range:2.0-3.1), and 3.7 for the sliding window segmentation method (range:2.6-5.0).
Conclusion:This study shows that while smart sequencing segmentation method significantly reduced the total number of monitor units it did not reduce the total number of segments per plan. The smart sequencing segmentation method also produced a higher global maximum dose in all the cases, but with no additional sparing of the relevant organs at risk.