Department of Medical Physics, Al-Neelain University, Khartoum, Sudan.
Department of Physics, College of Science, Princess Nourah bint Abdulrahman University, Riyadh, Saudi Arabia.
J Appl Clin Med Phys. 2022 Jul;23(7):e13630. doi: 10.1002/acm2.13630. Epub 2022 May 9.
Deep learning-based knowledge-based planning (KBP) methods have been introduced for radiotherapy dose distribution prediction to reduce the planning time and maintain consistent high-quality plans. This paper presents a novel KBP model using an attention-gating mechanism and a three-dimensional (3D) U-Net for intensity-modulated radiation therapy (IMRT) 3D dose distribution prediction in head-and-neck cancer.
A total of 340 head-and-neck cancer plans, representing the OpenKBP-2020 AAPM Grand Challenge data set, were used in this study. All patients were treated with the IMRT technique and a dose prescription of 70 Gy. The data set was randomly divided into 64%/16%/20% as training/validation/testing cohorts. An attention-gated 3D U-Net architecture model was developed to predict full 3D dose distribution. The developed model was trained using the mean-squared error loss function, Adam optimization algorithm, a learning rate of 0.001, 120 epochs, and batch size of 4. In addition, a baseline U-Net model was also similarly trained for comparison. The model performance was evaluated on the testing data set by comparing the generated dose distributions against the ground-truth dose distributions using dose statistics and clinical dosimetric indices. Its performance was also compared to the baseline model and the reported results of other deep learning-based dose prediction models.
The proposed attention-gated 3D U-Net model showed high capability in accurately predicting 3D dose distributions that closely replicated the ground-truth dose distributions of 68 plans in the test set. The average value of the mean absolute dose error was 2.972 ± 1.220 Gy (vs. 2.920 ± 1.476 Gy for a baseline U-Net) in the brainstem, 4.243 ± 1.791 Gy (vs. 4.530 ± 2.295 Gy for a baseline U-Net) in the left parotid, 4.622 ± 1.975 Gy (vs. 4.223 ± 1.816 Gy for a baseline U-Net) in the right parotid, 3.346 ± 1.198 Gy (vs. 2.958 ± 0.888 Gy for a baseline U-Net) in the spinal cord, 6.582 ± 3.748 Gy (vs. 5.114 ± 2.098 Gy for a baseline U-Net) in the esophagus, 4.756 ± 1.560 Gy (vs. 4.992 ± 2.030 Gy for a baseline U-Net) in the mandible, 4.501 ± 1.784 Gy (vs. 4.925 ± 2.347 Gy for a baseline U-Net) in the larynx, 2.494 ± 0.953 Gy (vs. 2.648 ± 1.247 Gy for a baseline U-Net) in the PTV_70, and 2.432 ± 2.272 Gy (vs. 2.811 ± 2.896 Gy for a baseline U-Net) in the body contour. The average difference in predicting the D value for the targets (PTV_70, PTV_63, and PTV_56) was 2.50 ± 1.77 Gy. For the organs at risk, the average difference in predicting the (brainstem, spinal cord, and mandible) and (left parotid, right parotid, esophagus, and larynx) values was 1.43 ± 1.01 and 2.44 ± 1.73 Gy, respectively. The average value of the homogeneity index was 7.99 ± 1.45 for the predicted plans versus 5.74 ± 2.95 for the ground-truth plans, whereas the average value of the conformity index was 0.63 ± 0.17 for the predicted plans versus 0.89 ± 0.19 for the ground-truth plans. The proposed model needs less than 5 s to predict a full 3D dose distribution of 64 × 64 × 64 voxels for a new patient that is sufficient for real-time applications.
The attention-gated 3D U-Net model demonstrated a capability in predicting accurate 3D dose distributions for head-and-neck IMRT plans with consistent quality. The prediction performance of the proposed model was overall superior to a baseline standard U-Net model, and it was also competitive to the performance of the best state-of-the-art dose prediction method reported in the literature. The proposed model could be used to obtain dose distributions for decision-making before planning, quality assurance of planning, and guiding-automated planning for improved plan consistency, quality, and planning efficiency.
深度学习为放疗剂量分布预测引入了基于知识的规划 (KBP) 方法，以减少规划时间并保持一致的高质量计划。本文提出了一种新颖的 KBP 模型，使用注意力门控机制和三维 (3D) U-Net 进行头颈部癌症调强放疗 (IMRT) 3D 剂量分布预测。
本研究使用了总共 340 个头颈部癌症计划，代表了 OpenKBP-2020 AAPM 大挑战数据集。所有患者均采用 IMRT 技术和 70 Gy 的剂量处方进行治疗。数据集随机分为 64%/16%/20%作为训练/验证/测试队列。开发了一种基于注意力门控的 3D U-Net 架构模型来预测全 3D 剂量分布。所开发的模型使用均方误差损失函数、Adam 优化算法、学习率为 0.001、120 个 epoch 和批量大小为 4 进行训练。此外，还类似地训练了一个基线 U-Net 模型进行比较。通过比较生成的剂量分布与地面实况剂量分布，使用剂量统计和临床剂量学指标来评估测试数据集上模型的性能。还将其性能与基线模型和文献中报道的其他基于深度学习的剂量预测模型的性能进行了比较。
所提出的注意力门控 3D U-Net 模型在准确预测 3D 剂量分布方面表现出了很高的能力，这些分布能够很好地复制测试集中 68 个计划的地面实况剂量分布。在脑干中，平均绝对剂量误差值为 2.972±1.220 Gy（与基线 U-Net 的 2.920±1.476 Gy 相比），左腮腺中为 4.243±1.791 Gy（与基线 U-Net 的 4.530±2.295 Gy 相比），右腮腺中为 4.622±1.975 Gy（与基线 U-Net 的 4.223±1.816 Gy 相比），脊髓中为 3.346±1.198 Gy（与基线 U-Net 的 2.958±0.888 Gy 相比），食管中为 6.582±3.748 Gy（与基线 U-Net 的 5.114±2.098 Gy 相比），下颌骨中为 4.756±1.560 Gy（与基线 U-Net 的 4.992±2.030 Gy 相比），喉部中为 4.501±1.784 Gy（与基线 U-Net 的 4.925±2.347 Gy 相比），PTV_70 中为 2.494±0.953 Gy（与基线 U-Net 的 2.648±1.247 Gy 相比），体轮廓中为 2.432±2.272 Gy（与基线 U-Net 的 2.811±2.896 Gy 相比）。预测靶区（PTV_70、PTV_63 和 PTV_56）D 值的平均差异为 2.50±1.77 Gy。对于危及器官，预测（脑干、脊髓和下颌骨）和（左腮腺、右腮腺、食管和喉部）值的平均差异分别为 1.43±1.01 和 2.44±1.73 Gy。预测计划的均匀性指数平均值为 7.99±1.45，而地面实况计划的均匀性指数平均值为 5.74±2.95，预测计划的适形性指数平均值为 0.63±0.17，而地面实况计划的适形性指数平均值为 0.89±0.19。所提出的模型可以在不到 5 秒的时间内预测新患者的 64×64×64 体素全 3D 剂量分布，足以满足实时应用的需求。
注意力门控 3D U-Net 模型在预测头颈部 IMRT 计划的准确 3D 剂量分布方面表现出一致的高质量能力。所提出模型的预测性能总体优于基线标准 U-Net 模型，并且与文献中报道的最佳基于深度学习的剂量预测方法的性能相当。所提出的模型可用于在计划前进行剂量分布决策、计划质量保证以及指导自动化计划，以提高计划的一致性、质量和效率。
