深度学习剂量预测模型在宫颈癌容积旋转调强放疗中的对比研究。

Deep learning (DL) is widely used in dose prediction for radiation oncology, multiple DL techniques comparison is often lacking in the literature. To compare the performance of 4 state-of-the-art DL models in predicting the voxel-level dose distribution for cervical cancer volumetric modulated arc therapy (VMAT). A total of 261 patients' plans for cervical cancer were retrieved in this retrospective study. A three-channel feature map, consisting of a planning target volume (PTV) mask, organs at risk (OARs) mask, and CT image was fed into the three-dimensional (3D) U-Net and its 3 variants models. The data set was randomly divided into 80% as training-validation and 20% as testing set, respectively. The model performance was evaluated on the 52 testing patients by comparing the generated dose distributions against the clinical approved ground truth (GT) using mean absolute error (MAE), dose map difference (GT-predicted), clinical dosimetric indices, and dice similarity coefficients (DSC). The 3D U-Net and its 3 variants DL models exhibited promising performance with a maximum MAE within the PTV 0.83% ± 0.67% in the UNETR model. The maximum MAE among the OARs is the left femoral head, which reached 6.95% ± 6.55%. For the body, the maximum MAE was observed in UNETR, which is 1.19 ± 0.86%, and the minimum MAE was 0.94 ± 0.85% for 3D U-Net. The average error of the Dmean difference for different OARs is within 2.5 Gy. The average error of V40 difference for the bladder and rectum is about 5%. The mean DSC under different isodose volumes was above 90%. DL models can predict the voxel-level dose distribution accurately for cervical cancer VMAT treatment plans. All models demonstrated almost analogous performance for voxel-wise dose prediction maps. Considering all voxels within the body, 3D U-Net showed the best performance. The state-of-the-art DL models are of great significance for further clinical applications of cervical cancer VMAT.

深度学习（DL）在放射肿瘤学中的剂量预测中得到了广泛应用，但文献中往往缺乏对多种 DL 技术的比较。本研究旨在比较 4 种最先进的 DL 模型在预测宫颈癌容积调强弧形治疗（VMAT）中体素级剂量分布的性能。

本回顾性研究共检索了 261 例宫颈癌患者的计划。一个三通道特征图，由一个计划靶区（PTV）掩模、危及器官（OARs）掩模和 CT 图像组成，被输入到三维（3D）U-Net 及其 3 个变体模型中。数据集随机分为 80%作为训练-验证集和 20%作为测试集。通过将生成的剂量分布与临床批准的真实值（GT）进行比较，使用平均绝对误差（MAE）、剂量图差异（GT-预测）、临床剂量学指标和骰子相似系数（DSC）评估模型在 52 例测试患者中的性能。3D U-Net 及其 3 个变体 DL 模型表现出良好的性能，在 UNETR 模型中，PTV 的最大 MAE 为 0.83%±0.67%。OARs 中最大的 MAE 是左侧股骨头，达到 6.95%±6.55%。对于身体，UNETR 中观察到的最大 MAE 为 1.19%±0.86%，3D U-Net 的最小 MAE 为 0.94%±0.85%。不同 OARs 的 Dmean 差异的平均误差在 2.5Gy 以内。膀胱和直肠的 V40 差异的平均误差约为 5%。不同等剂量体积下的平均 DSC 均高于 90%。

DL 模型可以准确预测宫颈癌 VMAT 治疗计划的体素级剂量分布。所有模型在体素剂量预测图上表现出几乎相同的性能。考虑到体内所有体素，3D U-Net 表现出最佳性能。最先进的 DL 模型对宫颈癌 VMAT 的进一步临床应用具有重要意义。