Guo Yi, Chen Jun, Lu Lin, Qiu Lingna, Lan Linzhen, Guo Feibao, Hong Jinsheng
Department of Radiation Therapy, Cancer Center, The First Affiliated Hospital of Fujian Medical University, Fuzhou, 350005, China.
Department of Radiotherapy, National Regional Medical Center, Binhai Campus of The First Affiliated Hospital, Fujian Medical University, Fuzhou, 350212, China.
Radiat Oncol. 2025 Feb 25;20(1):26. doi: 10.1186/s13014-025-02603-0.
Deformable registration plays an important role in the accurate delineation of tumors. Most of the existing deep learning methods ignored two issues that can lead to inaccurate registration, including the limited field of view in MR scans and the different scanning angles that can exist between multimodal images. The purpose of this study is to improve the registration accuracy between CT and MR for nasopharyngeal carcinoma cases.
269 cases were enrolled in the study, and 188 cases were designated for training, while a separate set of 81 cases was reserved for testing. Each case had a CT volume and a T1-MR volume. The treatment table was removed from their CT images. The CycleFCNs model was used for deformable registration, and two strategies including adaptive mask registration strategy and weight allocation strategy were adopted for training. Dice similarity coefficient, Hausdorff distance, precision, and recall were calculated for normal tissues of CT-MR image pairs, before and after the registration. Three deformable registration methods including RayStation, Elastix, and VoxelMorph were compared with the proposed method.
The registration results of RayStation and Elastix are essentially consistent. Upon employing the VoxelMorph model and the proposed method for registration, a clear trend of increased dice similarity coefficient and decreased hausdorff distance can be observed. It is noteworthy that for the temporomandibular joint, pituitary, optic nerve, and optic chiasma, the proposed method has improved the average dice similarity coefficient from 0.86 to 0.91, 0.87 to 0.93, 0.85 to 0.89, and 0.77 to 0.83, respectively, as compared to RayStation. Additionally, within the same anatomical structures, the average hausdorff distance has been decreased from 2.98 mm to 2.28 mm, 1.83 mm to 1.53 mm, 3.74 mm to 3.56 mm, and 5.94 mm to 5.87 mm. Compared to the original CycleFCNs model, the improved model has significantly enhanced the dice similarity coefficient of the brainstem, pituitary gland, and optic nerve (P < 0.001).
The proposed method significantly improved the registration accuracy for multi-modal medical images in NPC cases. These findings have important clinical implications, as increased registration accuracy can lead to more precise tumor segmentation, optimized treatment planning, and ultimately, improved patient outcomes.
可变形配准在肿瘤的精确勾画中起着重要作用。现有的大多数深度学习方法忽略了两个可能导致配准不准确的问题，包括磁共振扫描中的视野受限以及多模态图像之间可能存在的不同扫描角度。本研究的目的是提高鼻咽癌病例中CT与MR之间的配准精度。
本研究纳入269例病例，其中188例用于训练，另外81例单独留作测试。每个病例有一个CT容积和一个T1-MR容积。从其CT图像中去除治疗床。使用CycleFCNs模型进行可变形配准，并采用自适应掩码配准策略和权重分配策略两种策略进行训练。计算配准前后CT-MR图像对正常组织的骰子相似系数、豪斯多夫距离、精度和召回率。将包括RayStation、Elastix和VoxelMorph在内的三种可变形配准方法与所提出的方法进行比较。
RayStation和Elastix的配准结果基本一致。采用VoxelMorph模型和所提出的方法进行配准后，可以观察到骰子相似系数增加和豪斯多夫距离减小的明显趋势。值得注意的是，对于颞下颌关节、垂体、视神经和视交叉，与RayStation相比，所提出的方法分别将平均骰子相似系数从0.86提高到0.91、从0.87提高到0.93、从0.85提高到0.89、从0.77提高到0.83。此外，在相同的解剖结构内，平均豪斯多夫距离从2.98mm降至2.28mm、从1.83mm降至1.53mm、从3.74mm降至3.56mm、从5.94mm降至5.87mm。与原始的CycleFCNs模型相比，改进后的模型显著提高了脑干、垂体和视神经的骰子相似系数（P < 0.001）。
所提出的方法显著提高了鼻咽癌病例中多模态医学图像的配准精度。这些发现具有重要的临床意义，因为配准精度的提高可以导致更精确的肿瘤分割、优化的治疗计划，并最终改善患者的预后。
