Corpus ID: 235742963

Double-Uncertainty Assisted Spatial and Temporal Regularization Weighting for Learning-based Registration

  title={Double-Uncertainty Assisted Spatial and Temporal Regularization Weighting for Learning-based Registration},
  author={Zhe Xu and Jie Luo and Donghuan Lu and Jiangpeng Yan and Jayender Jagadeesan and William M. Wells and Sarah F. Frisken and Kai Ma and Yefeng Zheng and Raymond Kai-yu Tong},
In order to tackle the difficulty associated with the ill-posed nature of the image registration problem, researchers use regularization to constrain the solution space. For most learning-based registration approaches, the regularization usually has a fixed weight and only constrains the spatial transformation. Such convention has two limitations: (1) The regularization strength of a specific image pair should be associated with the content of the images, thus the “one value fits all” scheme is… Expand

Figures and Tables from this paper


Reliability-Driven, Spatially-Adaptive Regularization for Deformable Registration
A reliability measure is proposed that identifies informative image cues useful for registration, and a novel, data-driven approach to spatially adapt regularization to the local image content via use of the proposed measure is presented. Expand
Probabilistic non-linear registration with spatially adaptive regularisation
The results indicate that using the proposed spatially adaptive prior leads to sparser deformations, which provide better localisation of regional volume change, and the proposed regularisation model leads to more data driven and localised maps of registration uncertainty. Expand
Sparse Bayesian registration of medical images for self‐tuning of parameters and spatially adaptive parametrization of displacements
This work extends Bayesian models of non‐rigid image registration to allow not only for the automatic determination of registration parameters, but also for a data‐driven, multiscale, spatially adaptive parametrization of deformations, using an efficient Variational Bayes framework. Expand
A Cooperative Autoencoder for Population-Based Regularization of CNN Image Registration
A novel neural network architecture is proposed that simultaneously learns and uses the population-level statistics of the spatial transformations to regularize the neural networks for unsupervised image registration, in the form of a bottleneck autoencoder. Expand
Unimodal Cyclic Regularization For Training Multimodal Image Registration Networks
This work proposes a unimodal cyclic regularization training pipeline, which learns task-specific prior knowledge from simpler unimmodal registration, to constrain the deformation field of multimodal registration. Expand
Probabilistic inference of regularisation in non-rigid registration
A probabilistic registration framework that infers the level of regularisation from the data and shows that inferring regularisation on an individual basis leads to a reduction in model over-fitting as measured by image folding while providing a similar level of overlap. Expand
Bayesian characterization of uncertainty in intra-subject non-rigid registration
A Bayesian non-rigid registration framework where conventional dissimilarity and regularization energies can be included in the likelihood and the prior distribution on deformations respectively through the use of Boltzmann's distribution is proposed. Expand
An Unsupervised Learning Model for Deformable Medical Image Registration
The proposed method uses a spatial transform layer to reconstruct one image from another while imposing smoothness constraints on the registration field, and demonstrates registration accuracy comparable to state-of-the-art 3D image registration, while operating orders of magnitude faster in practice. Expand
Unsupervised Multimodal Image Registration with Adaptative Gradient Guidance
A novel multimodal registration framework is proposed, which leverages the deformation fields estimated from both: the original to-be-registered image pair, and their corresponding gradient intensity maps, and adaptively fuses them with the proposed gated fusion module. Expand
Unsupervised Learning for Fast Probabilistic Diffeomorphic Registration
This paper presents a probabilistic generative model and derive an unsupervised learning-based inference algorithm that makes use of recent developments in convolutional neural networks (CNNs) and results in state of the art accuracy and very fast runtimes, while providing diffeomorphic guarantees and uncertainty estimates. Expand