Related Experiment Video
Updated: Aug 6, 2026
Radiation Planning Assistant - A Streamlined, Fully Automated Radiotherapy Treatment Planning System
Published on: April 11, 2018
Gamma passing rates of sequential boost head and neck VMAT treatment plans: complexity metrics and irradiation
K Outsikas1, G Patatoukas2, M Chalkia2
1Department of Applied Medical Physics, Medical School, "Attikon" University Hospital, National and Kapodistrian University of Athens, Athens, Greece; Institute of Biophysics and Biomedical Engineering, Faculty of Science, University of Lisbon, Lisbon, Portugal.
Purpose:
To investigate parameters systematically associated with low Gamma Passing Rate (GPR) values in Volumetric Modulated Arc Therapy (VMAT) technique in head and neck radiotherapy plans and to evaluate whether complexity metrics can predict Patient-Specific Quality Assurance (PSQA) pass or fail status reducing PSQA workload.
Methods:
304 head and neck VMAT treatment plans were retrospectively analyzed. Institutional-specific tolerance limits (TL) and action limits (AL) were calculated for each irradiation phase. Plan complexity was assessed using metrics from literature and three novel composite metrics were proposed. Statistical comparisons between the PTV irradiation phases as well as between low- and high-GPR subgroups per PTV irradiation phase were performed. Correlation and ROC analysis were used to assess the relationship between metrics and GPR, and their ability to identify low-GPR plans.
Results:
GPR values decrease significantly with increasing irradiation phase. Complexity metrics showed weak to moderate correlations with GPR, with strength varying per irradiation phase. Composite metrics showed improved classification performance, yielding low False Positive Rate (FPR) and identifying significant percentage of low-GPR plans.
Conclusion:
Deliverability in head and neck VMAT plans strongly depends on the PTV irradiation phase, demonstrated by both the complexity metrics and GPR values. Composite metrics effectively improve the identification of treatment plans with high likelihood of PSQA failure. The results of this study support the use of irradiation-phase-specific TL and AL for clinical use and for the in-depth understanding of PSQA results. Finally, this study establishes a framework for the integration of complexity metrics into the PSQA workflow.
