Matkovic Luke, Lei Yang, Fu Yabo, Wang Tonghe, Kesarwala Aparna H, Axente Marian, Roper Justin, Higgins Kristin, Bradley Jeffrey D, Liu Tian, Yang Xiaofeng
Department of Radiation Oncology and Winship Cancer Institute, Emory University, Atlanta, Georgia, USA.
Department of Medical Physics, Memorial Sloan Kettering Cancer Center, New York, NY, USA.
Med Phys. 2024 Mar;51(3):1974-1984. doi: 10.1002/mp.16738. Epub 2023 Sep 14.
An automated, accurate, and efficient lung four-dimensional computed tomography (4DCT) image registration method is clinically important to quantify respiratory motion for optimal motion management.
The purpose of this work is to develop a weakly supervised deep learning method for 4DCT lung deformable image registration (DIR).
The landmark-driven cycle network is proposed as a deep learning platform that performs DIR of individual phase datasets in a simulation 4DCT. This proposed network comprises a generator and a discriminator. The generator accepts moving and target CTs as input and outputs the deformation vector fields (DVFs) to match the two CTs. It is optimized during both forward and backward paths to enhance the bi-directionality of DVF generation. Further, the landmarks are used to weakly supervise the generator network. Landmark-driven loss is used to guide the generator's training. The discriminator then judges the realism of the deformed CT to provide extra DVF regularization.
We performed four-fold cross-validation on 10 4DCT datasets from the public DIR-Lab dataset and a hold-out test on our clinic dataset, which included 50 4DCT datasets. The DIR-Lab dataset was used to evaluate the performance of the proposed method against other methods in the literature by calculating the DIR-Lab Target Registration Error (TRE). The proposed method outperformed other deep learning-based methods on the DIR-Lab datasets in terms of TRE. Bi-directional and landmark-driven loss were shown to be effective for obtaining high registration accuracy. The mean and standard deviation of TRE for the DIR-Lab datasets was 1.20 ± 0.72 mm and the mean absolute error (MAE) and structural similarity index (SSIM) for our datasets were 32.1 ± 11.6 HU and 0.979 ± 0.011, respectively.
The landmark-driven cycle network has been validated and tested for automatic deformable image registration of patients' lung 4DCTs with results comparable to or better than competing methods.
一种自动化、准确且高效的肺部四维计算机断层扫描（4DCT）图像配准方法对于量化呼吸运动以实现最佳运动管理具有重要的临床意义。
本研究旨在开发一种用于4DCT肺部可变形图像配准（DIR）的弱监督深度学习方法。
提出了地标驱动循环网络作为深度学习平台，用于在模拟4DCT中对各个相位数据集进行DIR。该网络由一个生成器和一个判别器组成。生成器接受移动CT和目标CT作为输入，并输出变形矢量场（DVF）以匹配这两个CT。在正向和反向路径中对其进行优化，以增强DVF生成的双向性。此外，地标用于对生成器网络进行弱监督。地标驱动损失用于指导生成器的训练。然后，判别器判断变形CT的真实性，以提供额外的DVF正则化。
我们对来自公共DIR-Lab数据集的10个4DCT数据集进行了四折交叉验证，并对我们的临床数据集（包括50个4DCT数据集）进行了留出测试。通过计算DIR-Lab目标配准误差（TRE），使用DIR-Lab数据集评估了所提出方法相对于文献中其他方法的性能。在所提出的方法在DIR-Lab数据集上的TRE方面优于其他基于深度学习的方法。双向和地标驱动损失被证明对于获得高配准精度是有效的。DIR-Lab数据集的TRE的平均值和标准差分别为1.20±0.72毫米，我们数据集的平均绝对误差（MAE）和结构相似性指数（SSIM）分别为32.1±11.6 HU和0.979±0.011。
地标驱动循环网络已针对患者肺部4DCT的自动可变形图像配准进行了验证和测试，其结果与竞争方法相当或更好。
