Service de Physique Médicale et de Radioprotection, Centre Intégré de Cancérologie, CHU de Québec-Université Laval et Centre de recherche du CHU de Québec, Quebec, Quebec, Canada.
Département de Physique, de Génie Physique et d'Optique et Centre de Recherche sur le Cancer, Université Laval, Quebec, Quebec, Canada.
Phys Med Biol. 2024 Apr 9;69(8). doi: 10.1088/1361-6560/ad3418.
In brachytherapy, deep learning (DL) algorithms have shown the capability of predicting 3D dose volumes. The reliability and accuracy of such methodologies remain under scrutiny for prospective clinical applications. This study aims to establish fast DL-based predictive dose algorithms for low-dose rate (LDR) prostate brachytherapy and to evaluate their uncertainty and stability.Data from 200 prostate patients, treated withI sources, was collected. The Monte Carlo (MC) ground truth dose volumes were calculated with TOPAS considering the interseed effects and an organ-based material assignment. Two 3D convolutional neural networks, UNet and ResUNet TSE, were trained using the patient geometry and the seed positions as the input data. The dataset was randomly split into training (150), validation (25) and test (25) sets. The aleatoric (associated with the input data) and epistemic (associated with the model) uncertainties of the DL models were assessed.For the full test set, with respect to the MC reference, the predicted prostatemetric had mean differences of -0.64% and 0.08% for the UNet and ResUNet TSE models, respectively. In voxel-by-voxel comparisons, the average global dose difference ratio in the [-1%, 1%] range included 91.0% and 93.0% of voxels for the UNet and the ResUNet TSE, respectively. One forward pass or prediction took 4 ms for a 3D dose volume of 2.56 M voxels (128 × 160 × 128). The ResUNet TSE model closely encoded the well-known physics of the problem as seen in a set of uncertainty maps. The ResUNet TSE rectum Dhad the largest uncertainty metric of 0.0042.The proposed DL models serve as rapid dose predictors that consider the patient anatomy and interseed attenuation effects. The derived uncertainty is interpretable, highlighting areas where DL models may struggle to provide accurate estimations. The uncertainty analysis offers a comprehensive evaluation tool for dose predictor model assessment.
在近距离放射治疗中，深度学习（DL）算法已经显示出预测三维剂量体积的能力。对于预期的临床应用，这些方法的可靠性和准确性仍在审查中。本研究旨在为低剂量率（LDR）前列腺近距离放射治疗建立快速基于深度学习的预测剂量算法，并评估其不确定性和稳定性。
收集了 200 名接受 I 源治疗的前列腺患者的数据。使用 TOPAS 考虑到种子间效应和基于器官的材料分配，计算了蒙特卡罗（MC）地面真实剂量体积。使用患者几何形状和种子位置作为输入数据，对两个 3D 卷积神经网络，即 UNet 和 ResUNet TSE，进行了训练。数据集随机分为训练集（150）、验证集（25）和测试集（25）。评估了 DL 模型的偶然（与输入数据相关）和认知（与模型相关）不确定性。
对于整个测试集，与 MC 参考相比，UNet 和 ResUNet TSE 模型的预测前列腺体积的平均差异分别为-0.64%和 0.08%。在体素对体素的比较中，在[-1%，1%]范围内的平均全局剂量差异比包含 91.0%和 93.0%的体素，分别用于 UNet 和 ResUNet TSE。对于 2.56 M 体素（128×160×128）的 3D 剂量体积，一次前向传递或预测需要 4ms。ResUNet TSE 模型紧密地编码了众所周知的问题物理，这可以从一组不确定性图中看出。ResUNet TSE 直肠 Dhad 的不确定性度量最大，为 0.0042。
提出的 DL 模型可作为快速剂量预测器，考虑患者解剖结构和种子间衰减效应。所得到的不确定性是可解释的，突出了 DL 模型可能难以提供准确估计的区域。不确定性分析为剂量预测器模型评估提供了全面的评估工具。
