Overview of the hybrid-QA method. (a) All treatment plans are verified with a model-based QA. Whenever a treatment plan fails to pass clinical acceptance criteria, the physician was consulted whether the dose differences were still clinically acceptable. (b) A random selection of approved treatment plans (approximately 15%) was evaluated with a measurement-based QA to assess the agreement between the model-based QA and measurement-based QA. (c) A standard head and neck IMRT treatment plan was evaluated monthly to verify linac stability for IMRT delivery.
Comparison of COMPASS, Pinnacle, and Monte Carlo calculated depth dose curves (top) and dose profiles (bottom) in a water phantom with a 2.5 cm diameter cylinder of 2.4 g/cm3 density at 10 cm depth (insets).
Comparison of COMPASS, Pinnacle, and Monte Carlo calculated depth dose curves in a lung phantom with densities of 0.25 (lung), 1.00 (tissue), and 1.20 (tumor) g/cm3 (inset). Secondary build-up effects are visible at −12 cm (entrance of “tumor”) and −2 cm (entrance of “mediastinum”).
GI for 330 treatment plans with 325 PASS classifications (GI ≤ 0.4) and five EVAL classifications (0.4 < GI < 0.6). No treatment plans were classified as FAIL (GI ≥ 0.6).
Model-based QA and measurement-based QA agreed for 41 treatment plans (●) and resulted in different classifications for seven treatment plans (▲). All 48 treatment plans were accepted for treatment.
Average work load per treatment plan in minutes based on four representative head and neck IMRT treatment plans if no problems occur and performed by an experienced user. Measurements for the four treatment plans were completed in one single measurement session.
Article metrics loading...
Full text loading...