Unsupervised Learning for Fast Probabilistic Diffeomorphic Registration

@inproceedings{Dalca2018UnsupervisedLF,
  title={Unsupervised Learning for Fast Probabilistic Diffeomorphic Registration},
  author={Adrian V. Dalca and Guha Balakrishnan and John V. Guttag and Mert Rory Sabuncu},
  booktitle={MICCAI},
  year={2018}
}
Traditional deformable registration techniques achieve impressive results and offer a rigorous theoretical treatment, but are computationally intensive since they solve an optimization problem for each image pair. Recently, learning-based methods have facilitated fast registration by learning spatial deformation functions. However, these approaches use restricted deformation models, require supervised labels, or do not guarantee a diffeomorphic (topology-preserving) registration. Furthermore… Expand
Unsupervised Learning of Probabilistic Diffeomorphic Registration for Images and Surfaces
TLDR
A probabilistic generative model is presented and an unsupervised learning-based inference algorithm is derived that uses insights from classical registration methods and makes use of recent developments in convolutional neural networks (CNNs). Expand
Learning a Probabilistic Model for Diffeomorphic Registration
TLDR
The unsupervised method uses a conditional variational autoencoder network and constrain transformations to be symmetric and diffeomorphic by applying a differentiable exponentiation layer with a symmetric loss function and provides multi-scale velocity field estimations. Expand
Improved Probabilistic Diffeomorphic Registration with CNNs
Classical image registration methods have seen significant technical development, but often require significant runtime. In contrast, learning-based registration methods achieve impressive testExpand
Conditional Deformable Image Registration with Convolutional Neural Network
TLDR
This paper proposes a conditional image registration method and a new selfsupervised learning paradigm for deep deformable image registration that demonstrates that optimal solutions with arbitrary hyperparameters can be captured by a single deep convolutional neural network. Expand
A novel unsupervised learning model for diffeomorphic image registration
Deformable medical image registration is vital for doctor's diagnosis and quantitative analysis. In this paper, we propose a novel unsupervised learning model (denoted as BSADM) for 3D diffeomorphicExpand
Unsupervised Deep-Learning Based Deformable Image Registration: A Bayesian Framework
TLDR
This work introduces a fully Bayesian framework for unsupervised DL-based deformable image registration and provides an estimate of the uncertainty in the deformation-field by characterizing the true posterior distribution, thus, avoiding potential over-fitting. Expand
Intensity-Based Wasserstein Distance As A Loss Measure For Unsupervised Deformable Deep Registration
TLDR
The sliced Wasserstein distance is considered a valuable metric for deep mono-modal and multi- modal deformable medical image registration problems with the proposed implementation of a VoxelMorph backbone architecture based on a combination of UNet and spatial transformer networks. Expand
Large Deformation Diffeomorphic Image Registration with Laplacian Pyramid Networks
TLDR
A deep Laplacian Pyramid Image Registration Network is proposed, which can solve the image registration optimization problem in a coarse-to-fine fashion within the space of diffeomorphic maps. Expand
VoxelMorph: A Learning Framework for Deformable Medical Image Registration
TLDR
VoxelMorph promises to speed up medical image analysis and processing pipelines while facilitating novel directions in learning-based registration and its applications and demonstrates that the unsupervised model’s accuracy is comparable to the state-of-the-art methods while operating orders of magnitude faster. Expand
Highly Accurate and Memory Efficient Unsupervised Learning-Based Discrete CT Registration Using 2.5D Displacement Search
TLDR
A highly accurate unsupervised learning framework for 3D abdominal CT registration that uses a discrete displacement layer and a contrast-invariant metric (MIND descriptors) that is evaluated in a probabilistic fashion is proposed. Expand
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 37 REFERENCES
An Unsupervised Learning Model for Deformable Medical Image Registration
TLDR
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
SVF-Net: Learning Deformable Image Registration Using Shape Matching
TLDR
An innovative approach for registration based on the deterministic prediction of the parameters from both images instead of the optimization of a energy criteria is proposed and shows an important improvement over a state of the art optimization based algorithm. Expand
Non-rigid image registration using fully convolutional networks with deep self-supervision
TLDR
A novel non-rigid image registration algorithm that is built upon fully convolutional networks (FCNs) to optimize and learn spatial transformations between pairs of images to be registered that has been evaluated for registering 3D structural brain magnetic resonance (MR) images and obtained better performance than state-of-the-art image registration algorithms. Expand
Non-rigid image registration using self-supervised fully convolutional networks without training data
  • Hongming Li, Yong Fan
  • Computer Science, Medicine
  • 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018)
  • 2018
TLDR
A novel non-rigid image registration algorithm built upon fully convolutional networks (FCNs) to optimize and learn spatial transformations between pairs of images to be registered in a self-supervised learning framework and has been evaluated for registering 3D structural brain magnetic resonance images and obtained better performance than state-of-the-art image registration algorithms. Expand
Nonrigid Image Registration Using Multi-scale 3D Convolutional Neural Networks
TLDR
The proposed RegNet is trained using a large set of artificially generated DVFs, does not explicitly define a dissimilarity metric, and integrates image content at multiple scales to equip the network with contextual information, thereby greatly simplifying the training problem. Expand
Quicksilver: Fast predictive image registration – A deep learning approach
TLDR
This paper introduces Quicksilver, a fast deformable image registration method that accurately predicts registrations obtained by numerical optimization, is very fast, achieves state-of-the-art registration results on four standard validation datasets, and can jointly learn an image similarity measure. Expand
Dense image registration through MRFs and efficient linear programming
TLDR
A novel and efficient approach to dense image registration, which does not require a derivative of the employed cost function is introduced, and efficient linear programming using the primal dual principles is considered to recover the lowest potential of the cost function. Expand
Frequency Diffeomorphisms for Efficient Image Registration
TLDR
A novel finite dimensional Fourier representation of diffeomorphic deformations based on the key fact that the high frequency components of a diffeomorphism remain stationary throughout the integration process when computing the deformation associated with smooth velocity fields is introduced. Expand
Patch-Based Discrete Registration of Clinical Brain Images
TLDR
A non-rigid registration method for aligning brain images acquired in clinical settings that outperforms state of the art registration algorithms and avoids catastrophic failures often caused by these images. Expand
End-to-End Unsupervised Deformable Image Registration with a Convolutional Neural Network
TLDR
The results demonstrate that registration with DIRNet is as accurate as a conventional deformable image registration method with short execution times. Expand
...
1
2
3
4
...