Department of Biomedical Engineering, University of California, Davis, California, USA.
Department of Radiation Oncology, University of California Davis Medical Center, Sacramento, California, USA.
Med Phys. 2022 May;49(5):3263-3277. doi: 10.1002/mp.15585. Epub 2022 Mar 14.
Image guidance is used to improve the accuracy of radiation therapy delivery but results in increased dose to patients. This is of particular concern in children who need be treated per Pediatric Image Gently Protocols due to long-term risks from radiation exposure. The purpose of this study is to design a deep neural network architecture and loss function for improving soft-tissue contrast and preserving small anatomical features in ultra-low-dose cone-beam CTs (CBCT) of head and neck cancer (HNC) imaging.
A 2D compound U-Net architecture (modified U-Net++) with different depths was proposed to enhance the network capability of capturing small-volume structures. A mask weighted loss function (Mask-Loss) was applied to enhance soft-tissue contrast. Fifty-five paired CBCT and CT images of HNC patients were retrospectively collected for network training and testing. The output enhanced CBCT images from the present study were evaluated with quantitative metrics including mean absolute error (MAE), signal-to-noise ratio (SNR), and structural similarity (SSIM), and compared with those from the previously proposed network architectures (U-Net and wide U-Net) using MAE loss functions. A visual assessment of ten selected structures in the enhanced CBCT images of each patient was performed to evaluate image quality improvement, blindly scored by an experienced radiation oncologist specialized in HN cancer.
All the enhanced CBCT images showed reduced artifactual distortion and image noise. U-Net++ outperformed the U-Net and wide U-Net in terms of MAE, contrast near structure boundaries, and small structures. The proposed Mask-Loss improved image contrast and accuracy of the soft-tissue regions. The enhanced CBCT images predicted by U-Net++ and Mask-Loss demonstrated improvement compared to the U-Net in terms of average MAE (52.41 vs 42.85 HU), SNR (14.14 vs 15.07 dB), and SSIM (0.84 vs 0.87), respectively ( , in all paired t-tests). The visual assessment showed that the proposed U-Net++ and Mask-Loss significantly improved original CBCTs ( ), compared to the U-Net and MAE loss.
The proposed network architecture and loss function effectively improved image quality in soft-tissue contrast, organ boundary, and small structure preservation for ultra-low-dose CBCT following Image Gently Protocol. This method has potential to provide sufficient anatomical representation on the enhanced CBCT images for accurate treatment delivery and potentially fast online-adaptive re-planning for HN cancer patients.
图像引导被用于提高放射治疗的准确性，但会增加患者的剂量。对于需要根据儿科图像轻柔协议进行治疗的儿童，这尤其令人关注，因为他们会面临长期的辐射暴露风险。本研究的目的是设计一种深度神经网络架构和损失函数，用于改善头颈部癌症（HNC）成像中超低剂量锥形束 CT（CBCT）的软组织对比度和保留小解剖结构。
提出了一种具有不同深度的二维复合 U-Net 架构（改进的 U-Net++），以增强网络捕捉小体积结构的能力。应用掩模加权损失函数（Mask-Loss）来增强软组织对比度。回顾性收集了 55 对 HNC 患者的 CBCT 和 CT 图像，用于网络训练和测试。使用 MAE 损失函数比较了从本研究中输出的增强 CBCT 图像与先前提出的网络架构（U-Net 和宽 U-Net）的定量指标，包括平均绝对误差（MAE）、信噪比（SNR）和结构相似性（SSIM）。对每位患者的增强 CBCT 图像中的十个选定结构进行了视觉评估，以评估图像质量的改善，由专门从事 HN 癌症的经验丰富的放射肿瘤学家进行盲法评分。
所有增强的 CBCT 图像均显示出减少的人为失真和图像噪声。U-Net++在 MAE、对比度接近结构边界和小结构方面优于 U-Net 和宽 U-Net。所提出的 Mask-Loss 提高了软组织区域的图像对比度和准确性。与 U-Net 相比，U-Net++和 Mask-Loss 预测的增强 CBCT 图像在平均 MAE（52.41 与 42.85 HU）、SNR（14.14 与 15.07 dB）和 SSIM（0.84 与 0.87）方面均有所改善（ ，在所有配对 t 检验中）。视觉评估显示，与 U-Net 和 MAE 损失相比，所提出的 U-Net++和 Mask-Loss 显著改善了原始 CBCT（ ）。
所提出的网络架构和损失函数有效地提高了超剂量图像轻柔协议下的软组织对比度、器官边界和小结构保留的图像质量。该方法有可能在增强的 CBCT 图像上提供足够的解剖表示，用于准确的治疗交付，并为 HN 癌症患者提供潜在的快速在线自适应重新规划。
