Chen Wen, Li Yimin, Yuan Nimu, Qi Jinyi, Dyer Brandon A, Sensoy Levent, Benedict Stanley H, Shang Lu, Rao Shyam, Rong Yi
Department of Radiation Oncology, Xiangya Hospital, Central South University, Changsha, China.
Department of Radiation Oncology, University of California Davis Medical Center, Sacramento, CA, United States.
Front Artif Intell. 2021 Feb 11;3:614384. doi: 10.3389/frai.2020.614384. eCollection 2020.
To assess image quality and uncertainty in organ-at-risk segmentation on cone beam computed tomography (CBCT) enhanced by deep-learning convolutional neural network (DCNN) for head and neck cancer. An in-house DCNN was trained using forty post-operative head and neck cancer patients with their planning CT and first-fraction CBCT images. Additional fifteen patients with repeat simulation CT (rCT) and CBCT scan taken on the same day (oCBCT) were used for validation and clinical utility assessment. Enhanced CBCT (eCBCT) images were generated from the oCBCT using the in-house DCNN. Quantitative imaging quality improvement was evaluated using HU accuracy, signal-to-noise-ratio (SNR), and structural similarity index measure (SSIM). Organs-at-risk (OARs) were delineated on o/eCBCT and compared with manual structures on the same day rCT. Contour accuracy was assessed using dice similarity coefficient (DSC), Hausdorff distance (HD), and center of mass (COM) displacement. Qualitative assessment of users' confidence in manual segmenting OARs was performed on both eCBCT and oCBCT by visual scoring. eCBCT organs-at-risk had significant improvement on mean pixel values, SNR ( < 0.05), and SSIM ( < 0.05) compared to oCBCT images. Mean DSC of eCBCT-to-rCT (0.83 ± 0.06) was higher than oCBCT-to-rCT (0.70 ± 0.13). Improvement was observed for mean HD of eCBCT-to-rCT (0.42 ± 0.13 cm) vs. oCBCT-to-rCT (0.72 ± 0.25 cm). Mean COM was less for eCBCT-to-rCT (0.28 ± 0.19 cm) comparing to oCBCT-to-rCT (0.44 ± 0.22 cm). Visual scores showed OAR segmentation was more accessible on eCBCT than oCBCT images. DCNN improved fast-scan low-dose CBCT in terms of the HU accuracy, image contrast, and OAR delineation accuracy, presenting potential of eCBCT for adaptive radiotherapy.
为评估深度学习卷积神经网络（DCNN）增强的锥束计算机断层扫描（CBCT）对头颈部癌危及器官分割的图像质量和不确定性。使用40例头颈部癌术后患者的计划CT和首次分次CBCT图像训练了一个内部DCNN。另外15例在同一天进行重复模拟CT（rCT）和CBCT扫描（oCBCT）的患者用于验证和临床效用评估。使用内部DCNN从oCBCT生成增强CBCT（eCBCT）图像。使用HU准确性、信噪比（SNR）和结构相似性指数测量（SSIM）评估定量成像质量的改善。在o/eCBCT上勾勒出危及器官（OARs），并与同一天rCT上的手动结构进行比较。使用骰子相似系数（DSC）、豪斯多夫距离（HD）和质心（COM）位移评估轮廓准确性。通过视觉评分对用户在eCBCT和oCBCT上手动分割OARs的信心进行定性评估。与oCBCT图像相比，eCBCT危及器官的平均像素值、SNR（<0.05）和SSIM（<0.05）有显著改善。eCBCT与rCT的平均DSC（0.83±0.06）高于oCBCT与rCT（0.70±0.13）。观察到eCBCT与rCT的平均HD（0.42±0.13 cm）相对于oCBCT与rCT（0.72±0.25 cm）有所改善。eCBCT与rCT的平均COM（0.28±0.19 cm）低于oCBCT与rCT（0.44±0.22 cm）。视觉评分显示，在eCBCT上进行OAR分割比在oCBCT图像上更容易。DCNN在HU准确性、图像对比度和OAR勾勒准确性方面改善了快速扫描低剂量CBCT，显示了eCBCT在自适应放疗中的潜力。
