Jinsoo Uh1, Chuang Wang2, Sahaja Acharya2, Matthew J Krasin2, Chia-Ho Hua2. 1. Department of Radiation Oncology, St. Jude Children's Research Hospital, Memphis, United States. Electronic address: jinsoo.uh@stjude.org. 2. Department of Radiation Oncology, St. Jude Children's Research Hospital, Memphis, United States.
Abstract
PURPOSE: To train a deep neural network for correcting abdominal and pelvic cone-beam computed tomography (CBCT) of children and young adults in the presence of diverse patient size, anatomic extent, and scan parameters. MATERIALS AND METHODS: Pretreatment CBCT and planning/repeat CT image pairs from 64 children and young adults treated with proton therapy (aged 1-23 years) were analyzed. To evaluate the impact of anatomic extent in CBCT and data size in the training data, we compared the performance of three cycle-consistent generative adversarial network models that were separately trained by three datasets comprising abdominal (n = 21), pelvic (n = 29), and combined abdominal-pelvic image pairs (n = 50), respectively. The maximum body width of each patient was normalized to a fixed width before training and model application to reduce the impact of variations in body size. The corrected CBCT images by the three models were comparatively evaluated against the repeat CT closest in time to the CBCT (median gap, 0 days; range, 0-6 days) in HU accuracy, estimated dose distribution, and proton range. RESULTS: The network model trained by the combined dataset significantly outperformed the abdomen and pelvis models in mean absolute HU error of the corrected CBCT from 14 testing patients (47 ± 7 HU versus 51 ± 8 HU; paired Wilcoxon signed-rank test, P < 0.01). The larger error (60 ± 7 HU) without the body-size normalization confirmed the efficacy of the preprocessing. The model trained with the combined dataset resulted in gamma passing rates of 98.5 ± 1.9% (2%/2 mm criterion) and the range (80% distal fall-off) differences from the reference within ±3 mm for 91.2 ± 11.5% beamlets. CONCLUSION: Combining data from adjacent anatomic sites and normalizing age-dependent body sizes in children and young adults were beneficial in training a neural network to accurately estimate proton dose from CBCT despite limited training data size and anatomic diversities.
PURPOSE: To train a deep neural network for correcting abdominal and pelvic cone-beam computed tomography (CBCT) of children and young adults in the presence of diverse patient size, anatomic extent, and scan parameters. MATERIALS AND METHODS: Pretreatment CBCT and planning/repeat CT image pairs from 64 children and young adults treated with proton therapy (aged 1-23 years) were analyzed. To evaluate the impact of anatomic extent in CBCT and data size in the training data, we compared the performance of three cycle-consistent generative adversarial network models that were separately trained by three datasets comprising abdominal (n = 21), pelvic (n = 29), and combined abdominal-pelvic image pairs (n = 50), respectively. The maximum body width of each patient was normalized to a fixed width before training and model application to reduce the impact of variations in body size. The corrected CBCT images by the three models were comparatively evaluated against the repeat CT closest in time to the CBCT (median gap, 0 days; range, 0-6 days) in HU accuracy, estimated dose distribution, and proton range. RESULTS: The network model trained by the combined dataset significantly outperformed the abdomen and pelvis models in mean absolute HU error of the corrected CBCT from 14 testing patients (47 ± 7 HU versus 51 ± 8 HU; paired Wilcoxon signed-rank test, P < 0.01). The larger error (60 ± 7 HU) without the body-size normalization confirmed the efficacy of the preprocessing. The model trained with the combined dataset resulted in gamma passing rates of 98.5 ± 1.9% (2%/2 mm criterion) and the range (80% distal fall-off) differences from the reference within ±3 mm for 91.2 ± 11.5% beamlets. CONCLUSION: Combining data from adjacent anatomic sites and normalizing age-dependent body sizes in children and young adults were beneficial in training a neural network to accurately estimate proton dose from CBCT despite limited training data size and anatomic diversities.