S2374
Physics - Quality assurance and auditing
ESTRO 2026
Results:
References: 1. Herring, N., Kalla, M., & Paterson, D.J. The autonomic nervous system and cardiac arrhythmias: Current concepts and emerging therapies. Nature Reviews Cardiology 2019;16(12), 707–726. 2. Shah R, Assis F, Alugubelli N, Okada DR, Cardoso R, Shivkumar K, et al. Cardiac sympathetic denervation for refractory ventricular arrhythmias in patients with structural heart disease: A systematic review. Heart Rhythm. 2019;16(10):1499-505. Keywords: Commissioning, ventricular arrhythmia, MR-guided Digital Poster 3062 Moving Beyond Conventional Gamma Analysis: A New Metric for VMAT Verification Based on Statistical process Control and error Curve Model wenli lu 1 , xin yi 1 , hanyin zhang 1 , furong lv 2 1 Department of Oncology, The First Affiliated Hospital of Chongqing Medical University, chongqing, China. 2 Radiological Imaging Department, The First Affiliated Hospital of Chongqing Medical University, chongqing, China Purpose/Objective: While gamma analysis remains the cornerstone for VMAT plan pre-treatment verification, its well- documented limitations—including poor correlation with clinical dose errors and low sensitivity to key mechanical inaccuracies—necessitate more robust methodologies. This study introduces and validates an integrated workflow combining Statistical Process Control (SPC) and Error Curve Modeling (ECM) to identify superior verification metrics and establish quantitative, clinically-relevant pass/fail criteria. Material/Methods: We retrospectively analyzed 107 clinical VMAT plans (69 nasopharyngeal, 38 cervical cancers). Using a Delta4 phantom, we introduced five classes of realistic errors into each plan... generating 1938 verification datasets. Each dataset was evaluated against 3%/2mm, 2%/2mm, and 1%/1mm criteria across multiple analysis regions: the conventional ≥ 10% threshold, a composite 'whole' clinical region-of- interest (ROI), and TG-53/ESTRO-inspired 'divide & conquer' sub-regions (HD, HG, MD, LD). Our analytical workflow comprised: 1) Spearman correlation and inter-group testing to screen metrics strongly associated with dose errors (DE%); 2) SPC to define statistical tolerance limits; 3) Gaussian-fitted ECMs to quantify error detection capability; and 4) ROC analysis on an independent test set (n=649) for final performance validation.
The whole-region metrics ( γ _whole, μ GI_whole) consistently demonstrated strong correlations with PTV DE% across all Gamma criteria (r > 0.5, p < 0.01), unlike conventional γ _10%. SPC-derived tolerance limits varied significantly by disease site and Gamma criterion, underscoring the need for context-specific thresholds (e.g., for 3%/2mm: γ _whole-NPC=0.6822, μ GI_whole-NPC=0.8105; γ _whole-UCC=0.6344, μ GI_whole-UCC=0.5278). Error curve models revealed that γ _whole and μ GI_whole in NPC were sensitive to systematic MLC errors >2mm and MU errors >3%, while μ GI_whole in UCC under 3%/2mm was sensitive to all tested errors. Crucially, for detecting clinically significant DE% ( ≥ 3%), the whole-region metrics vastly outperformed γ _10%, with AUCs of 0.947 ( γ _whole- NPC) and 0.912 ( γ _whole-UCC) compared to <0.71 for γ _10%. In independent validation, the optimal metrics achieved high detection rates (88.75-100%) and balanced F1-scores (84.84-85%). Conclusion: This study validates a novel SPC-ECM workflow that effectively overcomes the critical limitations of conventional gamma analysis. We conclusively demonstrate that metrics based on the entire clinical ROI (whole) are superior for detecting clinically relevant errors. The proposed framework provides a robust, data-driven foundation for establishing individualized and clinically grounded pass/fail standards, offering a readily implementable solution for enhancing the quality and safety of radiotherapy. Keywords: SPC-ECM workflow, gamma analysis, clinical ROI Multi-Center Standardization and Practical Recommendations for Quality Assurance of Thermometry in Superficial Hyperthermia in the Netherlands Abiram Kanagaratnam 1 , Remko Zweije 2,3 , Abdelali Ameziane 1 , Carolina A. Carrapiço - Seabra 1 , Michelle van Wieren 4 , Patrick V. Granton 1 , Paola C. Tello Valverde 2,3 , Anton Rink 1 , Marion Essers 4 , Rianne de Kroon - Oldenhof 2,3 , Gerard C. van Rhoon 1 , Lisca F. Wurfbain 2,3 , Martine Franckena 1 , Desiree H.J.G. van den Bongard 2,3 , Akke Bakker 5,6 , Sergio Curto 1 1 Department of Radiotherapy, Erasmus MC Cancer Institute, Rotterdam, Netherlands. 2 Department of Radiation Oncology, Amsterdam UMC, Amsterdam, Netherlands. 3 Cancer Treatment and Quality of Life / Cancer Biology and Immunology, Cancer Center Amsterdam, Amsterdam, Netherlands. 4 Department of Mini-Oral 3160
Made with FlippingBook - Share PDF online