Akademik Veri Yönetim Sistemi
iARTIST-International Advanced Radiotherapy Techniques Symposium, İstanbul, Türkiye, 12 - 13 Haziran 2026, ss.1, (Özet Bildiri)
Purpose: The aim of this study was to evaluate the clinical feasibility and reliability of a knowledge-based planning (KBP) model with respect to plan quality (PTV coverage and OAR dose constraints) by retrospectively using treatment planning system (TPS) data from patients who underwent VMAT-SBRT for lung lesions.
Materials-Methods: A RapidPlan™ (RP) model was developed using 40 high-quality VMAT-SBRT plans as a training dataset from patients with lung lesions <=5 cm, treated with total doses of 50 or 60 Gy delivered in 5 or 8 fractions. For validation, RP-optimized plans generated for 20 new lesions not included in the training set were compared with clinically approved plans created by an experienced medical physicist using manual planning (MP) for the same lesions. Comparisons were performed using dose–volume histogram (DVH) parameters for the PTV and OARs. Statistical analyses were conducted using SPSS v.24. The Wilcoxon signed-rank test was applied for paired comparisons. To our knowledge, this is the first study comparing RP and MP plans, including all ICRU Report 91 Level 2 reporting metrics for both target volumes and OARs.
Results: In the validation cohort, 100% of the RP plans were found to be clinically acceptable with respect to PTV and OAR DVH parameters. Compared with MP plans, RP plans provided significantly higher doses for PTV and IGTV D50% and Dmean, as well as for PTV D0.03cc. Conversely, MP plans exhibited significantly higher PTV D99% and Paddick conformity index (CI) values than those of RP plans. No significant differences were observed between RP and MP plans in terms of R100%, Paddick gradient index (GI), R50%, and D2cm. Regarding OARs, RP plans significantly reduced the Dmean of both the aorta and the main bronchus, whereas MP plans showed a significant dose reduction only for rib D0.03cc compared with RP. The monitor units (MU) and modulation factor were significantly higher in MP plans than in RP plans. Additionally, the radiotherapy planning time was reduced to approximately 30 minutes using RP.
Conclusion: The RP-based KBP model developed in our clinic for lung VMAT-SBRT enables generation of high-quality treatment plans and is not affected by intra- and inter-planner variability, thereby supporting its clinical applicability in lung SBRT. By reducing planning time, RP can improve clinical workflow efficiency. KBP-based RP planning emerges as a strategic approach that ensures sustainable quality improvement in modern radiotherapy clinics.