9 MLC error detection with EPIBeam for VMAT patient quality assurance

Autor: Philippe Boissard, C. Malet, Pascal Dupuis, A. Seneclauze, T. Baudier
Rok vydání: 2018
Předmět:
Zdroj: Physica Medica. 56:6
ISSN: 1120-1797
DOI: 10.1016/j.ejmp.2018.09.022
Popis: Introduction In recent years, the part of Volumetric Modulated Arc Therapy (VMAT) in external radiotherapy treatments has become more and more important. In fact, this technique allows to deliver the right dose to the tumor region while sparing surrounding organs. To get this complexity, the correct functioning of the Multi-Leaf Collimator (MLC) is particularly critical. Controls are recommended to ensure that the accelerator is able to deliver correctly each treatment plan. Electronic Portal Imaging Devices (EPIDs) may be useful to perform some of patient specific Quality Assurance (QA) [1] . This study proposes to evaluate if, associated with the software EPIBeam (Dosisoft®), they are able to detect MLC errors which could have a significant clinical impact. Methods Ten VMAT plans QA for patients treated for prostate cancer using a Synergy (Elekta®) accelerator were performed using two methods. The first method uses ArcCheck (SunNuclear®) and the second uses the EPID IView GT (Elekta®) associated with the EPIBeam software. Gamma analyses were used using global 3%/ 3 mm criterion. In addition, for each plan, MLC errors were introduced. All the leaves of the same bank were shift by 0.5, 1, 2 and 3 mm in the opening direction. The clinical impact of these errors has been evaluated in the MONACO planning software (Elekta®). These modified plans were checked in their turn. A control is considered satisfactory if more than 95% of the points have a gamma index greater than 1. Results For MLC error free treatment plans, EPIBeam achieves better gamma analysis results: mean = 99.8% [Min: 98.7%/ Max = 100%] versus 99% [Min: 98.1%/ Max: 99.7%] with ArcCheck. For errors less than or equal to 1 mm, all controls were within the tolerances for both modalities. For 2 mm errors, the number of processing plan within tolerances is 1/10 for ArcCheck versus 4/10 with EPIBeam. For 3 mm errors, no plan is within tolerances for the ArcCHECK while a plan passes for EPIbeam. It should be noted that the 2 and 3 mm errors induces significate changes on the dose distribution inside the patient. For example, an increase of the maximum rectal wall dose of 2.9% for 2 mm and 4.9% for 3 mm are observed. Conclusions With the 3%/3 mm criterion for gamma index analysis, less MLC errors were detected with EPIBEAM as with ArcCheck. However, by working on the analysis criteria, the error detection capability may be improved. It should be also interesting to combine EPIBeam with other controls such as log file analysis to have better detection efficiency while keeping time savings [2] .
Databáze: OpenAIRE
Externí odkaz: