Isotropic Total Variation Regularization of Displacements in Parametric Image Registration

Spatial regularization is essential in image registration, which is an ill-posed problem. Regularization can help to avoid both physically implausible displacement fields and local minima during optimization. Tikhonov regularization (squared $\ell _{2}$ -norm) is unable to correctly represent non-smooth displacement fields, that can, for example, occur at sliding interfaces in the thorax and abdomen in image time-series during respiration. In this paper, isotropic Total Variation (TV) regularization is used to enable accurate registration near such interfaces. We further develop the TV-regularization for parametric displacement fields and provide an efficient numerical solution scheme using the Alternating Directions Method of Multipliers (ADMM). The proposed method was successfully applied to four clinical databases which capture breathing motion, including CT lung and MR liver images. It provided accurate registration results for the whole volume. A key strength of our proposed method is that it does not depend on organ masks that are conventionally required by many algorithms to avoid errors at sliding interfaces. Furthermore, our method is robust to parameter selection, allowing the use of the same parameters for all tested databases. The average target registration error (TRE) of our method is superior (10% to 40%) to other techniques in the literature. It provides precise motion quantification and sliding detection with sub-pixel accuracy on the publicly available breathing motion databases (mean TREs of 0.95 mm for DIR 4D CT, 0.96 mm for DIR COPDgene, 0.91 mm for POPI databases).

[1]  Michael Brady,et al.  MRF-Based Deformable Registration and Ventilation Estimation of Lung CT , 2013, IEEE Transactions on Medical Imaging.

[2]  Gary E. Christensen,et al.  A Measure for Characterizing Sliding on Lung Boundaries , 2013, Annals of Biomedical Engineering.

[3]  Stephen P. Boyd,et al.  Distributed Optimization and Statistical Learning via the Alternating Direction Method of Multipliers , 2011, Found. Trends Mach. Learn..

[4]  Philippe C. Cattin,et al.  Discontinuity Preserving Registration of Abdominal MR Images with Apparent Sliding Organ Motion , 2011, Abdominal Imaging.

[5]  Simon Hermann Evaluation of Scan-Line Optimization for 3D Medical Image Registration , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition.

[6]  Laurent Risser,et al.  An implicit sliding-motion preserving regularisation via bilateral filtering for deformable image registration , 2014, Medical Image Anal..

[7]  Gábor Székely,et al.  Investigating anisotropic diffusion for the registration of abdominal MR images , 2013, 2013 IEEE 10th International Symposium on Biomedical Imaging.

[8]  PennecXavier,et al.  Iconic feature based nonrigid registration , 2003 .

[9]  François-Xavier Vialard,et al.  Piecewise-diffeomorphic image registration: Application to the motion estimation between 3D CT lung images with sliding conditions , 2013, Medical Image Anal..

[10]  Jan Rühaak,et al.  A fast and accurate parallel algorithm for non-linear image registration using Normalized Gradient fields , 2014, ISBI.

[11]  Thomas Guerrero,et al.  Computing global minimizers to a constrained B-spline image registration problem from optimal l1 perturbations to block match data. , 2014, Medical physics.

[12]  Horst Bischof,et al.  A Duality Based Algorithm for TV- L 1-Optical-Flow Image Registration , 2007, MICCAI.

[13]  René Werner,et al.  TV-L1-Based 3D Medical Image Registration with the Census Cost Function , 2013, PSIVT.

[14]  Michael Unser,et al.  Fast parametric elastic image registration , 2003, IEEE Trans. Image Process..

[15]  L. Rudin,et al.  Nonlinear total variation based noise removal algorithms , 1992 .

[16]  Daniel Rueckert,et al.  Temporal sparse free-form deformations , 2013, Medical Image Anal..

[17]  Jan Rühaak,et al.  Highly accurate fast lung CT registration , 2013, Medical Imaging.

[18]  Laurent Risser,et al.  Complex Lung Motion Estimation via Adaptive Bilateral Filtering of the Deformation Field , 2013, MICCAI.

[19]  Nicholas Ayache,et al.  Iconic feature based nonrigid registration: the PASHA algorithm , 2003, Comput. Vis. Image Underst..

[20]  Jan Rühaak,et al.  Combining Automatic Landmark Detection and Variational Methods for Lung CT Registration , 2013 .

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

[22]  J. Rojo,et al.  Least median of squares filtering of locally optimal point matches for compressible flow image registration , 2012, Physics in medicine and biology.

[23]  René Werner,et al.  High Accuracy Optical Flow for 3D Medical Image Registration Using the Census Cost Function , 2013, PSIVT.

[24]  Torsten Rohlfing,et al.  Image Similarity and Tissue Overlaps as Surrogates for Image Registration Accuracy: Widely Used but Unreliable , 2012, IEEE Transactions on Medical Imaging.

[25]  Nassir Navab,et al.  Dense image registration through MRFs and efficient linear programming , 2008, Medical Image Anal..

[26]  Puneet Singla,et al.  Evaluation of advanced Lukas–Kanade optical flow on thoracic 4D-CT , 2013, Journal of Clinical Monitoring and Computing.

[27]  Tony F. Chan,et al.  Color TV: total variation methods for restoration of vector-valued images , 1998, IEEE Trans. Image Process..

[28]  Martin Burger,et al.  A Hyperelastic Regularization Energy for Image Registration , 2013, SIAM J. Sci. Comput..

[29]  Orcun Goksel,et al.  Imaging of sliding visceral interfaces during breathing , 2016, 2016 IEEE 13th International Symposium on Biomedical Imaging (ISBI).

[30]  Heinz Handels,et al.  Estimating Large Lung Motion in COPD Patients by Symmetric Regularised Correspondence Fields , 2015, MICCAI.

[31]  Terry M. Peters,et al.  RANCOR: Non-Linear Image Registration with Total Variation Regularization , 2014, ArXiv.

[32]  Mila Nikolova,et al.  Algorithms for Finding Global Minimizers of Image Segmentation and Denoising Models , 2006, SIAM J. Appl. Math..

[33]  Tobias Gass,et al.  Total Variation Regularization of Displacements in Parametric Image Registration , 2014, ABDI@MICCAI.

[34]  Eric A. Hoffman,et al.  Automatic lung segmentation for accurate quantitation of volumetric X-ray CT images , 2001, IEEE Transactions on Medical Imaging.

[35]  R. Castillo,et al.  Four-dimensional deformable image registration using trajectory modeling , 2010, Physics in medicine and biology.

[36]  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.

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

[38]  X. Pennec,et al.  3D non-rigid registration by gradient descent on a Gaussian-windowed similarity measure using convolutions , 2000, Proceedings IEEE Workshop on Mathematical Methods in Biomedical Image Analysis. MMBIA-2000 (Cat. No.PR00737).

[39]  D Sarrut,et al.  Registration of sliding objects using direction dependent B-splines decomposition , 2013, Physics in medicine and biology.

[40]  Michael J. Black,et al.  Secrets of optical flow estimation and their principles , 2010, 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[41]  José M. Bioucas-Dias,et al.  Algorithms for imaging inverse problems under sparsity regularization , 2012, 2012 3rd International Workshop on Cognitive Information Processing (CIP).