mlVIRNET: Multilevel Variational Image Registration Network

We present a novel multilevel approach for deep learning based image registration. Recently published deep learning based registration methods have shown promising results for a wide range of tasks. However, these algorithms are still limited to relatively small deformations. Our method addresses this shortcoming by introducing a multilevel framework, which computes deformation fields on different scales, similar to conventional methods. Thereby, a coarse-level alignment is obtained first, which is subsequently improved on finer levels. We demonstrate our method on the complex task of inhale-to-exhale lung registration. We show that the use of a deep learning multilevel approach leads to significantly better registration results.

[1]  Jan Modersitzky,et al.  FAIR - Flexible Algorithms for Image Registration , 2009, Fundamentals of algorithms.

[2]  Ruzena Bajcsy,et al.  Multiresolution elastic matching , 1989, Comput. Vis. Graph. Image Process..

[3]  Mert R. Sabuncu,et al.  An Unsupervised Learning Model for Deformable Medical Image Registration , 2018, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[4]  E. Regan,et al.  Genetic Epidemiology of COPD (COPDGene) Study Design , 2011, COPD.

[5]  Josien P. W. Pluim,et al.  Progressively growing convolutional networks for end-to-end deformable image registration , 2019, Medical Imaging: Image Processing.

[6]  Eldad Haber,et al.  Cofir: Coarse and Fine Image Registration , 2004 .

[7]  Yoshua Bengio,et al.  Understanding the difficulty of training deep feedforward neural networks , 2010, AISTATS.

[8]  Thomas Guerrero,et al.  A reference dataset for deformable image registration spatial accuracy evaluation using the COPDgene study archive , 2013, Physics in medicine and biology.

[9]  Max A. Viergever,et al.  A deep learning framework for unsupervised affine and deformable image registration , 2018, Medical Image Anal..

[10]  Mert R. Sabuncu,et al.  VoxelMorph: A Learning Framework for Deformable Medical Image Registration , 2018, IEEE Transactions on Medical Imaging.

[11]  Stefan Heldmann,et al.  Estimation of Large Motion in Lung CT by Integrating Regularized Keypoint Correspondences into Dense Deformable Registration , 2017, IEEE Transactions on Medical Imaging.

[12]  Maxime Sermesant,et al.  SVF-Net: Learning Deformable Image Registration Using Shape Matching , 2017, MICCAI.

[13]  Stefan Heldmann,et al.  Unsupervised learning for large motion thoracic CT follow-up registration , 2019, Image Processing.

[14]  Stefan Heldmann,et al.  Enhancing Label-Driven Deep Deformable Image Registration with Local Distance Metrics for State-of-the-Art Cardiac Motion Tracking , 2018, Bildverarbeitung für die Medizin.

[15]  Haiying Liu,et al.  A Generic Framework for Non-rigid Registration Based on Non-uniform Multi-level Free-Form Deformations , 2001, MICCAI.

[16]  Marc Modat,et al.  Label-driven weakly-supervised learning for multimodal deformarle image registration , 2017, 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018).