State Key Laboratory of Radio Frequency Heterogeneous Integration, Shenzhen University, Shenzhen 518060, Guangdong Province, People's Republic of China.
The Second People's Hospital of Futian District, Shenzhen 518049, Guangdong Province, People's Republic of China.
Phys Med Biol. 2024 Mar 28;69(7). doi: 10.1088/1361-6560/ad2a96.
This study aims to enhance medical image registration by addressing the limitations of existing approaches that rely on spatial transformations through U-Net, ConvNets, or Transformers. The objective is to develop a novel architecture that combines ConvNets, graph neural networks (GNNs), and capsule networks to improve the accuracy and efficiency of medical image registration, which can also deal with the problem of rotating registration.We propose an deep learning-based approach which can be utilized in both unsupervised and semi-supervised manners, named as HGCMorph. It leverages a hybrid framework that integrates ConvNets and GNNs to capture lower-level features, specifically short-range attention, while also utilizing capsule networks (CapsNets) to model abstract higher-level features, including entity properties such as position, size, orientation, deformation, and texture. This hybrid framework aims to provide a comprehensive representation of anatomical structures and their spatial relationships in medical images.The results demonstrate the superiority of HGCMorph over existing state-of-the-art deep learning-based methods in both qualitative and quantitative evaluations. In unsupervised training process, our model outperforms the recent SOTA method TransMorph by achieving 7%/38% increase on Dice score coefficient (DSC), and 2%/7% improvement on negative jacobian determinant for OASIS and LPBA40 datasets, respectively. Furthermore, HGCMorph achieves improved registration accuracy in semi-supervised training process. In addition, when dealing with complex 3D rotations and secondary randomly deformations, our method still achieves the best performance. We also tested our methods on lung datasets, such as Japanese Society of Radiology, Montgoermy and Shenzhen.The significance lies in its innovative design to medical image registration. HGCMorph offers a novel framework that overcomes the limitations of existing methods by efficiently capturing both local and abstract features, leading to enhanced registration accuracy, discontinuity-preserving, and pose-learning abilities. The incorporation of capsule networks introduces valuable improvements, making the proposed method a valuable contribution to the field of medical image analysis. HGCMorph not only advances the SOTA methods but also has the potential to improve various medical applications that rely on accurate image registration.
这项研究旨在通过解决现有方法的局限性来增强医学图像配准，这些方法依赖于 U-Net、ConvNets 或 Transformers 进行空间变换。目标是开发一种新的架构，该架构结合了 ConvNets、图神经网络（GNNs）和胶囊网络，以提高医学图像配准的准确性和效率，同时也能解决旋转配准的问题。
我们提出了一种基于深度学习的方法，可以在无监督和半监督两种情况下使用，名为 HGCMorph。它利用了一种混合框架，该框架结合了 ConvNets 和 GNNs 来捕获低级特征，特别是短程注意力，同时还利用胶囊网络（CapsNets）来对高级抽象特征进行建模，包括位置、大小、方向、变形和纹理等实体属性。这种混合框架旨在为医学图像中的解剖结构及其空间关系提供全面的表示。
实验结果表明，在定性和定量评估中，HGCMorph 优于现有的基于深度学习的方法。在无监督训练过程中，我们的模型在 Dice 得分系数（DSC）上分别比最近的 SOTA 方法 TransMorph 提高了 7%/38%，在 OASIS 和 LPBA40 数据集上的负雅可比行列式分别提高了 2%/7%。此外，HGCMorph 在半监督训练过程中实现了更高的注册精度。此外，当处理复杂的 3D 旋转和二次随机变形时，我们的方法仍然能够达到最佳性能。我们还在肺部数据集（如日本放射学会、蒙哥马利和深圳）上测试了我们的方法。
这项研究的意义在于其对医学图像配准的创新设计。HGCMorph 提供了一种新颖的框架，通过有效地捕获局部和抽象特征，克服了现有方法的局限性，从而提高了配准精度、保持不连续性和学习姿势的能力。胶囊网络的引入带来了有价值的改进，使所提出的方法成为医学图像分析领域的一项有价值的贡献。HGCMorph 不仅推进了 SOTA 方法，还有潜力改善依赖于准确图像配准的各种医学应用。
