Liu, Chihray

Abstract Objective . UNet-based deep-learning (DL) architectures are promising dose engines for traditional linear accelerator (Linac) models. Current UNet-based engines, however, were designed differently with various strategies, making it challenging to fairly compare the results from different studies. The objective of this study is to thoroughly evaluate the performance of UNet-based models on magnetic-resonance (MR)-Linac-based intensity-modulated radiation therapy (IMRT) dose calculations. Approach . The UNet-based models, including the standard-UNet, cascaded-UNet, dense-dilated-UNet, residual-UNet, HD-UNet, and attention-aware-UNet, were implemented. The model input is patient CT and IMRT field dose in water, and the output is patient dose calculated by DL model. The reference dose was calculated by the Monaco Monte Carlo module. Twenty training and ten test cases of prostate patients were included. The accuracy of the DL-calculated doses was measured using gamma analysis, and the calculation efficiency was evaluated by inference time. Results . All the studied models effectively corrected low-accuracy doses in water to high-accuracy patient doses in a magnetic field. The gamma passing rates between reference and DL-calculated doses were over 86% (1%/1 mm), 98% (2%/2 mm), and 99% (3%/3 mm) for all the models. The inference times ranged from 0.03 (graphics processing unit) to 7.5 (central processing unit) seconds. Each model demonstrated different strengths in calculation accuracy and efficiency; Res-UNet achieved the highest accuracy, HD-UNet offered high accuracy with the fewest parameters but the longest inference, dense-dilated-UNet was consistently accurate regardless of model levels, standard-UNet had the shortest inference but relatively lower accuracy, and the others showed average performance. Therefore, the best-performing model would depend on the specific clinical needs and available computational resources. Significance . The feasibility of using common UNet-based models for MR-Linac-based dose calculations has been explored in this study. By using the same model input type, patient training data, and computing environment, a fair assessment of the models’ performance was present.

The purpose of this study is to investigate changes in lung tumor internal target volume during stereotactic body radiotherapy treatment (SBRT) using magnetic resonance imaging (MRI). Ten lung cancer patients (13 tumors) undergoing SBRT (48 Gy over four consecutive days) were evaluated. Each patient underwent three lung MRI evaluations: before SBRT (MRI‐1), after fraction 3 of SBRT (MRI‐3), and three months after completion of SBRT (MRI‐3m). Each MRI consisted of T1‐weighted images in axial plane through the entire lung. A cone‐beam CT (CBCT) was taken before each fraction. On MRI and CBCT taken before fractions 1 and 3, gross tumor volume (GTV) was contoured and differences between the two volumes were compared. Median tumor size on CBCT before fractions 1 (CBCT‐1) and 3 (CBCT‐3) was 8.68 and 11.10 cm3, respectively. In 12 tumors, the GTV was larger on CBCT‐3 compared to CBCT‐1 (median enlargement, 1.56 cm3). Median tumor size on MRI‐1, MRI‐3, and MRI‐3m was 7.91, 11.60, and 3.33 cm3, respectively. In all patients, the GTV was larger on MRI‐3 compared to MRI‐1 (median enlargement, 1.54 cm3). In all patients, GTV was smaller on MRI‐3m compared to MRI‐1 (median shrinkage, 5.44 cm3). On CBCT and MRI, all patients showed enlargement of the GTV during the treatment week of SBRT, except for one patient who showed minimal shrinkage (0.86 cm3). Changes in tumor volume are unpredictable; therefore, motion and breathing must be taken into account during treatment planning, and image‐guided methods should be used, when treating with large fraction sizes.PACS number: 87.53.Ly