Department of Integrated Health Science, Nagoya University Graduate School of Medicine, Aichi, Japan; Department of Medical Technology, Nagoya University Hospital, Aichi, Japan.
Takasaki Advanced Radiation Research Institute, Quantum Beam Science Research Directorate, National Institutes for Quantum Science and Technology (QST), Gunma, Japan.
Phys Med. 2022 Jul;99:130-139. doi: 10.1016/j.ejmp.2022.05.013. Epub 2022 Jun 9.
Proton-induced secondary-electron-bremsstrahlung (SEB) imaging is a promising method for estimating the ranges of particle beam. However, SEB images do not directly represent dose distributions of particle beams. In addition, the ranges estimated from measured images were deviated because of limited spatial resolutions of the developed x-ray camera as well as statistical noise in the images. To solve these problems, we proposed a method for predicting high-resolution dose images from SEB images with various count level using a deep learning (DL) approach for range and width verification.
In this study, we adopted the double U-Net model, which is a previously proposed deep convolutional network model. The first U-Net model in the double U-Net model was used to denoise the SEB images with various count level. The first U-Net model for denoising was trained on 8000 pairs of SEB images with various count level and noise-free images which were created by a sophisticated in-house developed model function. The second U-Net model for dose prediction was trained using 8000 pairs of denoised SEB images from the first U-Net model and high-resolution dose images generated by Monte Carlo simulation.
For both simulation and measurement data, the trained DL model could successfully predict high-resolution dose images which showed a clear Bragg peak and no statistical noise. The difference of the range and width was less than 2.1 mm, even from the SEB images measured with a decrease in the number of irradiated protons to less than 11% of 3.2 × 10 protons.
High-resolution dose images from measured and simulated SEB images were successfully predicted by using the trained DL model for protons. Our proposed DL model was feasible to predict dose images accurately even with smaller number of irradiated protons.
质子诱导的次级电子韧致辐射（SEB）成像是一种很有前途的方法，可以估计粒子束的射程。然而，SEB 图像并不能直接代表粒子束的剂量分布。此外，由于开发的 X 射线相机的空间分辨率有限以及图像中的统计噪声，从测量图像估计的射程会产生偏差。为了解决这些问题，我们提出了一种使用深度学习（DL）方法从具有各种计数水平的 SEB 图像预测高分辨率剂量图像的方法，用于射程和宽度验证。
在这项研究中，我们采用了双 U-Net 模型，这是一种以前提出的深度卷积网络模型。双 U-Net 模型中的第一个 U-Net 模型用于对具有各种计数水平的 SEB 图像进行去噪。第一个 U-Net 模型的去噪是在 8000 对具有各种计数水平和无噪声的 SEB 图像和由复杂的内部开发的模型函数创建的无噪声图像上进行训练的。第二个 U-Net 模型的剂量预测是使用 8000 对来自第一个 U-Net 模型的去噪 SEB 图像和通过蒙特卡罗模拟生成的高分辨率剂量图像进行训练的。
对于模拟和测量数据，经过训练的 DL 模型都可以成功预测高分辨率剂量图像，这些图像显示出清晰的布拉格峰，没有统计噪声。即使从照射质子数量减少到 3.2×10 个质子的 11%以下的 SEB 图像进行测量，射程和宽度的差异也小于 2.1mm。
使用训练有素的 DL 模型成功地从测量和模拟的 SEB 图像预测了高分辨率剂量图像。即使照射的质子数量较少，我们提出的 DL 模型也可以准确地预测剂量图像。
