Fast 4D elastic group-wise image registration. Convolutional interpolation revisited.

BACKGROUND AND OBJECTIVE This paper proposes a new and highly efficient implementation of 3D+t groupwise registration based on the free-form deformation paradigm. METHODS Deformation is posed as a cascade of 1D convolutions, achieving great reduction in execution time for evaluation of transformations and gradients. RESULTS The proposed method has been applied to 4D cardiac MRI and 4D thoracic CT monomodal datasets. Results show an average runtime reduction above 90%, both in CPU and GPU executions, compared with the classical tensor product formulation. CONCLUSIONS Our implementation, although fully developed for the metric sum of squared differences, can be extended to other metrics and its adaptation to multiresolution strategies is straightforward. Therefore, it can be extremely useful to speed up image registration procedures in different applications where high dimensional data are involved.

[1]  Daniel Rueckert,et al.  Nonrigid registration using free-form deformations: application to breast MR images , 1999, IEEE Transactions on Medical Imaging.

[2]  Sébastien Ourselin,et al.  Fast free-form deformation using graphics processing units , 2010, Comput. Methods Programs Biomed..

[3]  Patrick Clarysse,et al.  Spatiotemporal motion estimation for respiratory-correlated imaging of the lungs. , 2010, Medical physics.

[4]  Youbing Yin,et al.  Efficient methods for implementation of multi-level nonrigid mass-preserving image registration on GPUs and multi-threaded CPUs , 2016, Comput. Methods Programs Biomed..

[5]  Torsten Rohlfing,et al.  Nonrigid image registration in shared-memory multiprocessor environments with application to brains, breasts, and bees , 2003, IEEE Transactions on Information Technology in Biomedicine.

[6]  Marcos Martín-Fernández,et al.  Nonrigid groupwise registration for motion estimation and compensation in compressed sensing reconstruction of breath‐hold cardiac cine MRI , 2016, Magnetic resonance in medicine.

[7]  Santiago Aja-Fernández,et al.  Groupwise Elastic Registration by a New Sparsity-Promoting Metric: Application to the Alignment of Cardiac Magnetic Resonance Perfusion Images , 2013, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[8]  Stefan Klein,et al.  Fast Multidimensional B-spline Interpolation Using Template Metaprogramming , 2014, WBIR.

[9]  Alejandro F. Frangi,et al.  Multiresolution eXtended Free‐Form Deformations (XFFD) for non‐rigid registration with discontinuous transforms , 2017, Medical Image Anal..

[10]  Stefan Klein,et al.  Intrasubject multimodal groupwise registration with the conditional template entropy , 2018, Medical Image Anal..

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

[12]  R. Keys.  Cubic convolution interpolation for digital image processing , 1981 .

[13]  Marcos Martín-Fernández,et al.  An Efficient Multi-resolution Reconstruction Scheme with Motion Compensation for 5D Free-Breathing Whole-Heart MRI , 2017, CMMI/RAMBO/SWITCH@MICCAI.

[14]  Dinggang Shen,et al.  Deep Learning Deformation Initialization for Rapid Groupwise Registration of Inhomogeneous Image Populations , 2019, Frontiers in Neuroinformatics.

[15]  Paul Suetens,et al.  GPU-accelerated elastic 3D image registration for intra-surgical applications , 2011, Comput. Methods Programs Biomed..

[16]  D K Smith,et al.  Numerical Optimization , 2001, J. Oper. Res. Soc..

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

[18]  Gábor Székely,et al.  Parallel Mutual Information Based 3 D Non-Rigid Registration on a Multi-Core Platform , 2008 .

[19]  Thomas W. Sederberg,et al.  Free-form deformation of solid geometric models , 1986, SIGGRAPH.

[20]  Yuanjie Zheng,et al.  Deep Group-Wise Registration for Multi-Spectral Images From Fundus Images , 2019, IEEE Access.

[21]  Paul Suetens,et al.  Efficient GPU-accelerated elastic image registration , 2008 .

[22]  Alan V. Oppenheim,et al.  Discrete-Time Signal Pro-cessing , 1989 .

[23]  Xuhao Chen,et al.  Escoin: Efficient Sparse Convolutional Neural Network Inference on GPUs , 2018, 1802.10280.

[24]  Hsieh Hou,et al.  Cubic splines for image interpolation and digital filtering , 1978 .

[25]  M. Salman Asif,et al.  Motion‐adaptive spatio‐temporal regularization for accelerated dynamic MRI , 2013, Magnetic resonance in medicine.

[26]  Song Wang,et al.  A Parallel Nonrigid Registration Algorithm Based on B-Spline for Medical Images , 2016, Comput. Math. Methods Medicine.

[27]  Peng Xue,et al.  Lung 4D CT Image Registration Based on High-Order Markov Random Field , 2020, IEEE Transactions on Medical Imaging.

[28]  C. R. Deboor,et al.  A practical guide to splines , 1978 .

[29]  C. M. Reeves,et al.  Function minimization by conjugate gradients , 1964, Comput. J..

[30]  Stefan Klein,et al.  Free-Form Deformation Using Lower-Order B-spline for Nonrigid Image Registration , 2014, MICCAI.

[31]  Jeffrey A. Fessler,et al.  A Simple Regularizer for B-spline Nonrigid Image Registration That Encourages Local Invertibility , 2009, IEEE Journal of Selected Topics in Signal Processing.

[32]  Sung Yong Shin,et al.  Scattered Data Interpolation with Multilevel B-Splines , 1997, IEEE Trans. Vis. Comput. Graph..

[33]  W SederbergThomas,et al.  Free-form deformation of solid geometric models , 1986 .

[34]  Stefan Klein,et al.  Nonrigid registration of dynamic medical imaging data using nD + t B-splines and a groupwise optimization approach , 2011, Medical Image Anal..

[35]  Kumaradevan Punithakumar,et al.  A GPU-Accelerated Deformable Image Registration Algorithm With Applications to Right Ventricular Segmentation , 2017, IEEE Access.

[36]  Thibaud Briand,et al.  Theory and Practice of Image B-Spline Interpolation , 2018, Image Process. Line.

[37]  Marcos Martín-Fernández,et al.  Jacobian weighted temporal total variation for motion compensated compressed sensing reconstruction of dynamic MRI , 2017, Magnetic resonance in medicine.

[38]  J. Sarvaiya,et al.  Image Registration by Template Matching Using Normalized Cross-Correlation , 2009, 2009 International Conference on Advances in Computing, Control, and Telecommunication Technologies.

[39]  Santiago Aja-Fernández,et al.  Robust estimation of the apparent diffusion coefficient invariant to acquisition noise and physiological motion. , 2018, Magnetic resonance imaging.

[40]  Fumihiko Ino,et al.  A data distributed parallel algorithm for nonrigid image registration , 2005, Parallel Comput..

[41]  Santiago Aja-Fernández,et al.  Joint groupwise registration and ADC estimation in the liver using a B-value weighted metric. , 2018, Magnetic resonance imaging.

[42]  Wiro J Niessen,et al.  Groupwise image registration based on a total correlation dissimilarity measure for quantitative MRI and dynamic imaging data , 2018, Scientific Reports.

[43]  Pablo Casaseca-de-la-Higuera,et al.  Reconstruction techniques for cardiac cine MRI , 2019, Insights into Imaging.

[44]  Daniel Rueckert,et al.  Non-rigid registration using free-form deformations , 2015 .

[45]  Mert R. Sabuncu,et al.  Unsupervised Learning of Probabilistic Diffeomorphic Registration for Images and Surfaces , 2019, Medical Image Anal..