基于深度学习的头颈部 3D 剂量分布预测的知识型规划方法的对比研究。

A comparative study of deep learning-based knowledge-based planning methods for 3D dose distribution prediction of head and neck.

机构信息

Department of Medical Physics, Al-Neelain University, Khartoum, Sudan.

Department of Physics, College of Science, Princess Nourah bint Abdulrahman University, Riyadh, Saudi Arabia.

出版信息

J Appl Clin Med Phys. 2023 Sep;24(9):e14015. doi: 10.1002/acm2.14015. Epub 2023 May 3.

DOI:10.1002/acm2.14015

PMID:

Abstract

PURPOSE

In this paper, we compare four novel knowledge-based planning (KBP) algorithms using deep learning to predict three-dimensional (3D) dose distributions of head and neck plans using the same patients' dataset and quantitative assessment metrics.

METHODS

A dataset of 340 oropharyngeal cancer patients treated with intensity-modulated radiation therapy was used in this study, which represents the AAPM OpenKBP - 2020 Grand Challenge dataset. Four 3D convolutional neural network architectures were built. The models were trained on 64% of the data set and validated on 16% for voxel-wise dose predictions: U-Net, attention U-Net, residual U-Net (Res U-Net), and attention Res U-Net. The trained models were then evaluated for their performance on a test data set (20% of the data) by comparing the predicted dose distributions against the ground-truth using dose statistics and dose-volume indices.

RESULTS

The four KBP dose prediction models exhibited promising performance with an averaged mean absolute dose error within the body contour <3 Gy on 68 plans in the test set. The average difference in predicting the D index for all targets was 0.92 Gy (p = 0.51) for attention Res U-Net, 0.94 Gy (p = 0.40) for Res U-Net, 2.94 Gy (p = 0.09) for attention U-Net, and 3.51 Gy (p = 0.08) for U-Net. For the OARs, the values for the and indices were 2.72 Gy (p < 0.01) for attention Res U-Net, 2.94 Gy (p < 0.01) for Res U-Net, 1.10 Gy (p < 0.01) for attention U-Net, 0.84 Gy (p < 0.29) for U-Net.

CONCLUSION

All models demonstrated almost comparable performance for voxel-wise dose prediction. KBP models that employ 3D U-Net architecture as a base could be deployed for clinical use to improve cancer patient treatment by creating plans with consistent quality and making the radiotherapy workflow more efficient.

摘要

目的

本文使用深度学习比较了四种新的基于知识的规划（KBP）算法，以使用相同患者数据集和定量评估指标来预测头颈部计划的三维（3D）剂量分布。

方法

本研究使用了 340 名接受调强放射治疗的口咽癌患者的数据集，该数据集代表 AAPM OpenKBP-2020 挑战赛数据集。构建了四个 3D 卷积神经网络架构。该模型在数据集的 64%上进行训练，并在 16%上进行验证，以进行体素剂量预测：U-Net、注意力 U-Net、残差 U-Net（Res U-Net）和注意力 Res U-Net。然后，通过将预测的剂量分布与ground-truth 进行比较，使用剂量统计量和剂量体积指数评估在测试数据集（数据集的 20%）上的模型性能。

结果

在测试集中，68 个计划中，四个 KBP 剂量预测模型的表现均很有前景，体内轮廓的平均绝对剂量误差均<3Gy。预测所有靶区 D 指数的平均差异分别为注意力 Res U-Net 为 0.92Gy（p=0.51），Res U-Net 为 0.94Gy（p=0.40），注意力 U-Net 为 2.94Gy（p=0.09），U-Net 为 3.51Gy（p=0.08）。对于 OARs，指数和指数的值分别为注意力 Res U-Net 为 2.72Gy（p<0.01），Res U-Net 为 2.94Gy（p<0.01），注意力 U-Net 为 1.10Gy（p<0.01），U-Net 为 0.84Gy（p<0.29）。