Abstract:Background Deformable image registration is an essential technique of medical image analysis, which plays important roles in several clinical applications. Existing deep learning‐based registration methods have already achieved promising performance for the registrations with small deformations, while it is still challenging to deal with the large deformation registration due to the limits of the image intensity‐similarity‐based objective function. Purpose To achieve the image registration with large‐scale deformations, we proposed a multilevel network architecture FCNet to gradually refine the registration results based on semantic feature consistency constraint and flow normalization (FN) strategy. Methods At each level of FCNet, the architecture is mainly composed to a FeaExtractor, a FN module, and a spatial transformation module. FeaExtractor consists of three parallel streams which are used to extract the individual features of fixed and moving images, as well as their joint features, respectively. Using these features, the initial deformation field is estimated, which passes through a FN module to refine the deformation field based on the difference map of deformation filed between two adjacent levels. This allows the FCNet to progressively improve the registration performance. Finally, a spatial transformation module is used to get the warped image based on the deformation field. Moreover, in addition to the image intensity‐similarity‐based objective function, a semantic‐feature consistency constraint is also introduced, which can further promote the alignments by imposing the similarity between the fixed and warped image features. To validate the effectiveness of the proposed method, we compared our method with the state‐of‐the‐art methods on three different datasets. In EMPIRE10 dataset, 20, 3, and 7 fixed and moving 3D computer tomography (CT) image pairs were used for training, validation, and testing respectively; in IXI dataset, atlas to individual image registration task was performed, with 3D MR images of 408, 58, and 115 individuals were used for training, validation, and testing respectively; in the in‐house dataset, patient to atlas registration task was implemented, with the 3D MR images of 94, 3, and 15 individuals being training, validation, and testing sets, respectively. Results The qualitative and quantitative comparison results demonstrated that the proposed method is beneficial for handling large deformation image registration problems, with the DSC and ASSD improved by at least 1.0% and 25.9% on EMPIRE10 dataset. The ablation experiments also verified the effectiveness of the proposed feature combination strategy, feature consistency constraint, and FN module. Conclusions Our proposed FCNet enables multiscale registration from coarse to fine, surpassing existing SOTA registration methods and effectively handling long‐range spatial relationships.

mlVIRNET: Multilevel Variational Image Registration Network

Learning a Model-Driven Variational Network for Deformable Image Registration

A Deep Learning Framework for Unsupervised Affine and Deformable Image Registration

Linear and Deformable Image Registration with 3D Convolutional Neural Networks

MvMM-RegNet: A new image registration framework based on multivariate mixture model and neural network estimation

MICDIR: Multi-scale Inverse-consistent Deformable Image Registration using UNetMSS with Self-Constructing Graph Latent

MrRegNet: Multi-resolution Mask Guided Convolutional Neural Network for Medical Image Registration with Large Deformations

Voxelmorph++ Going beyond the cranial vault with keypoint supervision and multi-channel instance optimisation

Networks for Joint Affine and Non-parametric Image Registration

Deep learning-based 3D brain multimodal medical image registration

SearchMorph:Multi-scale Correlation Iterative Network for Deformable Registration

HCS-Net: Multi-level deformation strategy combined with quadruple attention for image registration

CNN-based Lung CT Registration with Multiple Anatomical Constraints

Residual Aligner Network

Multilevel network for large deformation image registration based on feature consistency and flow normalization

DELR-Net: a network for 3D multimodal medical image registration in more lightweight application scenarios

F3RNet: Full-Resolution Residual Registration Network for Deformable Image Registration

Multiscale unsupervised network for deformable image registration

Residual Aligner-based Network (RAN): Motion-separable structure for coarse-to-fine discontinuous deformable registration

Unsupervised deep learning for geometric feature detection and multilevel-multimodal image registration

L2NLF: a novel linear-to-nonlinear framework for multi-modal medical image registration