Multimodality image registration in the head-and-neck using a deep learning-derived synthetic CT as a bridge
Multimodality image registration in the head-and-neck using a deep learning-derived synthetic CT as a bridge
复制标题
DOI：
10.1002/mp.13976
复制
发表时间：
2020-01-02
期刊：
影响因子：
3.8
通讯作者：
Sheng, Ke
中科院分区：
文献类型：
--
作者：
McKenzie, Elizabeth M.;Santhanam, Anand;Sheng, Ke
目的：开发并证明使用基于深度学习的跨模态合成的新型头颈部多模态图像配准技术的有效性。方法25例头颈部肿瘤患者在同一天、同一制动状态下接受磁共振（MR）和计算机断层扫描（CT）。对所有MR-CT对使用五重交叉验证来训练神经网络以从MR图像生成合成CT。25例患者中有24例还进行了单独的CT检查（未固定），并用于测试。将CT非对准变形到合成CT，并与CT非对准配准到MR进行比较。从MR到CT非对准以及从合成CT到CT非对准进行相同的配准。所有配准均使用B样条对变形进行建模，并使用互信息对目标进行建模。结果进行了评估，使用95%的Hausdorff距离脊髓轮廓，地标误差，逆一致性，估计变形场的雅可比行列式。结果：当初始刚性错位较大时，将CT与MRI衍生合成CT配准比直接配准更好地对准脊髓。平均界标误差从MR -> CT非对准的9.8 +/- 3.1 mm降低到CTsynth -> CT非对准的变形配准的6.0 +/- 2.1 mm。在CT到MR方向上，标志误差从CT非对齐-> MR可变形配准中的10.0 +/- 4.3 mm降低到CT非对齐-> CTsynth可变形配准中的6.6 +/- 2.0 mm。雅可比行列式的平均值为0.98。所提出的方法也证明了改进的逆一致性的直接方法。结论我们表明，使用深度学习衍生的合成CT代替MR进行MR -> CT和CT -> MR变形配准，可提供优于直接多模态配准的上级结果。
Purpose To develop and demonstrate the efficacy of a novel head-and-neck multimodality image registration technique using deep-learning-based cross-modality synthesis. Methods and Materials Twenty-five head-and-neck patients received magnetic resonance (MR) and computed tomography (CT) (CTaligned) scans on the same day with the same immobilization. Fivefold cross validation was used with all of the MR-CT pairs to train a neural network to generate synthetic CTs from MR images. Twenty-four of 25 patients also had a separate CT without immobilization (CTnon-aligned) and were used for testing. CTnon-aligned's were deformed to the synthetic CT, and compared to CTnon-aligned registered to MR. The same registrations were performed from MR to CTnon-aligned and from synthetic CT to CTnon-aligned. All registrations used B-splines for modeling the deformation, and mutual information for the objective. Results were evaluated using the 95% Hausdorff distance among spinal cord contours, landmark error, inverse consistency, and Jacobian determinant of the estimated deformation fields. Results When large initial rigid misalignment is present, registering CT to MRI-derived synthetic CT aligns the cord better than a direct registration. The average landmark error decreased from 9.8 +/- 3.1 mm in MR -> CTnon-aligned to 6.0 +/- 2.1 mm in CTsynth -> CTnon-aligned deformable registrations. In the CT to MR direction, the landmark error decreased from 10.0 +/- 4.3 mm in CTnon-aligned -> MR deformable registrations to 6.6 +/- 2.0 mm in CTnon-aligned -> CTsynth deformable registrations. The Jacobian determinant had an average value of 0.98. The proposed method also demonstrated improved inverse consistency over the direct method. Conclusions We showed that using a deep learning-derived synthetic CT in lieu of an MR for MR -> CT and CT -> MR deformable registration offers superior results to direct multimodal registration.