S2369
Physics - Quality assurance and auditing
ESTRO 2026
clinical feasibility. Material/Methods:
introduced shifts with an accuracy always better than 0.5 mm (mean 0.2 mm) independent from lesion size but with less dosimetric accuracy (see figure).
A total of 23 clinical VMAT treatment plans, optimized in Pinnacle TPS and previously verified using Delta4®, were retrospectively analyzed with LINACWatch®. The dataset included treatments of the breast, thorax, pelvis, head and neck, and liver treated at Radiotherapy Department of Emicenter, Naples, Italy.LINACWatch® reconstructs the delivered fluence from linear accelerator logfiles by analyzing dynamic parameters such as MLC leaf trajectories, gantry and collimator angles, dose rate, and monitor units. Validation was performed through a dual comparison approach: first, phantom-based QA measurements obtained with Delta4® were compared against the corresponding TPS-dose distributions, serving as reference standard; second, QA outcomes reconstructed by LINACWatch® were compared with the same TPS plans.Agreement between methods was quantified using γ Passing Rate (GPR). Confusion matrix was generated to compute Accuracy, Error Rate, Recall, Precision, and F-score, providing an objective evaluation of LINACWatch®’s error-detection capability and its potential integration into routine, data-driven QA workflows. Results: The analysis of 23 VMAT plans revealed a strong concordance between LINACWatch® and Delta4®. The confusion matrix (Fig.1) showed 22 True Positives, 1 False Negative, and no False Positives and True Negatives, yielding an Accuracy and Recall of 95.65%, an Error Rate of 4.35%, a Precision of 100%, and an F- score of 97.8%.
Conclusion: Relevant geometric misalignments ≤ 1 mm for SIMT- SRS could be detected reliably by film dosimetry whereas point dose measurements are only sensitive to larger shifts depending on lesion size. By combining both methods our new methodology gives trustworthy information about dosimetric as well as geometric accuracy. Keywords: stereotactic radiosurgery, radiochromic films Digital Poster 2841 Revolutionizing Radiotherapy QA: Real-Time Error Detection with LINACWatch® Gaetano Gagliardo 1,2 , Caterina Muto 1,2 , Marcello Serra 3 , Ferdinando Francomacaro 1,2 , Ivona Zlatkova 1,2 , Vincenzo Iorio 1,2 , Paolo Muto 1 1 Radiotherapy, Fondazione Muto ETS, Naples, Italy. 2 Radiotherapy, Emicenter, Naples, Italy. 3 Radiotherapy, IRCCS Fondazione "G. Pascale", Naples, Italy Purpose/Objective: With the increasing complexity of IMRT and VMAT treatments, radiotherapy quality assurance (QA) requires innovative, data-driven approaches that move beyond traditional phantom-based measurements. This study evaluates LINACWatch®, a logfile-based platform for real-time beam delivery verification, benchmarking its performance against the Delta4® system to assess detection accuracy and
Figure 1: Confusion matrix obtained from the comparative analysis of LINACWatch® and Delta4® with respect to the TPS reference. The performance of LINACWatch®-TPS was assessed using the γ Passing Rate under the 1%/1 mm criterion with a 95th- percentile acceptance limit, while Delta4®-TPS evaluations applied the 2%/2 mm criterion and a 90th- percentile threshold.The system accurately detected subtle delivery deviations-such as transient MLC misalignments and gantry speed fluctuations- confirming its sensitivity to clinically relevant errors.Beyond its accuracy, the logfile-based workflow required no additional irradiation and considerably
Made with FlippingBook - Share PDF online