Calibrated uncertainty estimation for interpretable proton computed tomography image correction using Bayesian deep learning
Calibrated uncertainty estimation for interpretable proton computed tomography image correction using Bayesian deep learning
复制标题
DOI：
10.1088/1361-6560/abe956
复制
发表时间：
2021-02
影响因子：
3.5
通讯作者：
Y. Nomura;Sodai Tanaka;Jeff Wang;H. Shirato;S. Shimizu;L. Xing
中科院分区：
文献类型：
--
作者：
Y. Nomura;Sodai Tanaka;Jeff Wang;H. Shirato;S. Shimizu;L. Xing
集成型质子计算机断层扫描（pCT）测量质子阻止功率比（SPR）图像用于质子治疗计划，但其图像质量因噪声和散射而降低。尽管已经提出了几种校正方法，但包括不确定性估计的技术是有限的。该研究提出了一种新的不确定性感知的pCT图像校正方法，使用贝叶斯卷积神经网络（BCNN）。构建了一个基于DenseNet的BCNN来预测校正后的SPR图像及其来自噪声SPR图像的不确定性。用Monte Carlo模拟方法收集了6个非拟人模型和3个头部模型的432幅有噪声的SPR图像，而真正的无噪声图像是用已知的几何和化学成分计算的。通过训练25个独特的BCNN模型，进行异方差损失和深度集成技术来估计任意和认知的不确定性。200-对每个模型独立地进行历元端到端训练。通过两次事后校准和计算特定点的光程长度不确定度分布，证明了预测不确定度的可行性。为了评估，头部SPR图像的准确性和水等效厚度（WET）校正训练的BCNN模型与传统的方法和非贝叶斯CNN模型进行了比较。BCNN校正的SPR图像代表具有高精度的无噪声图像。测试数据的平均绝对误差从未校正图像的0.263改善到BCNN校正图像的0.0538。此外，校准的不确定度代表准确的置信水平，BCNN校正的校准WET比非贝叶斯CNN更准确，具有高统计显著性。使用消费级GPU计算一个图像及其25个BCNN模型的不确定性的计算时间为0.7秒。我们的模型能够预测准确的pCT图像以及两种类型的不确定性。这些不确定性将有助于识别SPR误差的潜在原因，并制定一个特定点的范围边界标准，对不确定性引导的质子治疗的阐述。
Integrated-type proton computed tomography (pCT) measures proton stopping power ratio (SPR) images for proton therapy treatment planning, but its image quality is degraded due to noise and scatter. Although several correction methods have been proposed, techniques that include estimation of uncertainty are limited. This study proposes a novel uncertainty-aware pCT image correction method using a Bayesian convolutional neural network (BCNN). A DenseNet-based BCNN was constructed to predict both a corrected SPR image and its uncertainty from a noisy SPR image. A total 432 noisy SPR images of 6 non-anthropomorphic and 3 head phantoms were collected with Monte Carlo simulations, while true noise-free images were calculated with known geometric and chemical components. Heteroscedastic loss and deep ensemble techniques were performed to estimate aleatoric and epistemic uncertainties by training 25 unique BCNN models. 200-epoch end-to-end training was performed for each model independently. Feasibility of the predicted uncertainty was demonstrated after applying two post-hoc calibrations and calculating spot-specific path length uncertainty distribution. For evaluation, accuracy of head SPR images and water-equivalent thickness (WET) corrected by the trained BCNN models was compared with a conventional method and non-Bayesian CNN model. BCNN-corrected SPR images represent noise-free images with high accuracy. Mean absolute error in test data was improved from 0.263 for uncorrected images to 0.0538 for BCNN-corrected images. Moreover, the calibrated uncertainty represents accurate confidence levels, and the BCNN-corrected calibrated WET was more accurate than non-Bayesian CNN with high statistical significance. Computation time for calculating one image and its uncertainties with 25 BCNN models is 0.7 s with a consumer grade GPU. Our model is able to predict accurate pCT images as well as two types of uncertainty. These uncertainties will be useful to identify potential cause of SPR errors and develop a spot-specific range margin criterion, toward elaboration of uncertainty-guided proton therapy.