American Institute of Mathematical Sciences

March  2021, 14(3): 1145-1160. doi: 10.3934/dcdss.2020386

Signed-distance function based non-rigid registration of image series with varying image intensity

 a. Department of Mathematics, FNSPE, Czech Technical University in Prague, Trojanova 13,120 00 Prague, Czech Republic b. Department of Radiology, Institute for clinical and experimental medicine, Vídeňská 1958/9, Praha 4,140 21, Czech Republic c. Inria, France d. LMS, Ecole Polytechnique, CNRS, Institut Polytechnique de Paris, France e. School of Biomedical Engineering & Imaging Sciences, St Thomas' Hospital, King's College London, UK

* Corresponding author: katerina.skardova@fjfi.cvut.cz

Received  January 2019 Revised  February 2020 Published  March 2021 Early access  June 2020

In this paper we propose a method for locally adjusted optical flow-based registration of multimodal images, which uses the segmentation of object of interest and its representation by the signed-distance function (OF$^{dist}$ method). We deal with non-rigid registration of the image series acquired by the Modiffied Look-Locker Inversion Recovery (MOLLI) magnetic resonance imaging sequence, which is used for a pixel-wise estimation of $T_1$ relaxation time. The spatial registration of the images within the series is necessary to compensate the patient's imperfect breath-holding. The evolution of intensities and a large variation of image contrast within the MOLLI image series, together with the myocardium of left ventricle (the object of interest) typically not being the most distinct object in the scene, makes the registration challenging. The paper describes all components of the proposed OF$^{dist}$ method and their implementation. The method is then compared to the performance of a standard mutual information maximization-based registration method, applied either to the original image (MIM) or to the signed-distance function (MIM$^{dist}$). Several experiments with synthetic and real MOLLI images are carried out. On synthetic image with a single object, MIM performed the best, while OF$^{dist}$ and MIM$^{dist}$ provided better results on synthetic images with more than one object and on real images. When applied to signed-distance function of two objects of interest, MIM$^{dist}$ provided a larger registration error, but more homogeneously distributed, compared to OF$^{dist}$. For the real MOLLI image series with left ventricle pre-segmented using a level-set method, the proposed OF$^{dist}$ registration performed the best, as is demonstrated visually and by measuring the increase of mutual information in the object of interest and its neighborhood.

Citation: Kateřina Škardová, Tomáš Oberhuber, Jaroslav Tintěra, Radomír Chabiniok. Signed-distance function based non-rigid registration of image series with varying image intensity. Discrete & Continuous Dynamical Systems - S, 2021, 14 (3) : 1145-1160. doi: 10.3934/dcdss.2020386
References:

show all references

References:
Diagram of the proposed method
Volume elements and the central points
Source image (A), target image (B), the absolute value of their difference before registration (C) and after registration by OF$^{dist}$, MIM and MIM$^{dist}$ (D-F). Parameters of OF$^{dist}$: $N_1 = 200, N_2 = 200, \alpha = 1.0, \beta = 3.25, \gamma = 2.5$
Source image (A), target image (B) and the absolute value of their difference before (C) and after registration by OF, MIM and MIM$^{dist}$ (D-F). Parameters of OF$^{dist}$: $N_1 = 200, N_2 = 200, \alpha = 1.5, \beta = 3.75, \gamma = 3.25$
Absolute value of the difference between the target and source signed-distance function before and after registration by OF$^{dist}$ and MIM$^{dist}$. The signed-distance functions are computed on 10-pixel-wide neighborhood of the edges of the object and set to constant outside the neighborhood
The results of OF$^{dist}$ registration of object marked by green line in 6a, and global MIM registration of the whole scene
Images from the MOLLI sequence with segmented myocardium. Parameters for the outer edge detection: $K_{S_1} = 1.3\cdot 10^{-6}$, $K_{S_2} = 1.3\cdot 10^{-6}$, $K_{T} = 2.3\cdot 10^{-6}$. Parameters for the inner edge detection: $K_{S_1} = K_{S_2} = K_{T} = 9.0\cdot 10^{-6}$
Results of registration of $S_1,S_2$ with target image $T$. Parameters of OF$^{dist}$: $N_1 = 256, N_2 = 218, \alpha = 1.25, \beta = 3.5, \gamma = 3.0$
Norms of difference between the target 3b image and source image 3a before and after registration by OF$^{dist}$, MIM and MIM$^{dist}$
 $\| T - S \|_2$ $\| T - S_{OF^{dist}} \|_2$ $\| T - S_{MIM} \|_2$ $\| T - S_{MIM^{dist}} \|_2$ 9462.47 3130.83 2173.93 2607.66
 $\| T - S \|_2$ $\| T - S_{OF^{dist}} \|_2$ $\| T - S_{MIM} \|_2$ $\| T - S_{MIM^{dist}} \|_2$ 9462.47 3130.83 2173.93 2607.66
Norms of difference between the target image in Figure 4b image and source image in Figure 4a before and after registration by OF$^{dist}$, MIM and MIM$^{dist}$
 $\| T - S \|_2$ $\| T - S_{OF^{dist}} \|_2$ $\| T - S_{MIM} \|_2$ $\| T - S_{MIM^{dist}} \|_2$ 6654.44 2199.07 4121.95 3602.27
 $\| T - S \|_2$ $\| T - S_{OF^{dist}} \|_2$ $\| T - S_{MIM} \|_2$ $\| T - S_{MIM^{dist}} \|_2$ 6654.44 2199.07 4121.95 3602.27
Norms of difference between target and source signed-distance function before and after registration by OF$^{dist}$ and MIM$^{dist}$. The source and target objects can be seen in Figure 4a and 4b, respectively
 $\| \phi_T - \phi_S \|_2$ $\| \phi_T - \phi_{S,{OF^{dist}}} \|_2$ $\| \phi_T - \phi_{S,{MIM^{dist}}} \|_2$ 0.849915 0.239367 0.2950385
 $\| \phi_T - \phi_S \|_2$ $\| \phi_T - \phi_{S,{OF^{dist}}} \|_2$ $\| \phi_T - \phi_{S,{MIM^{dist}}} \|_2$ 0.849915 0.239367 0.2950385
Norms of difference between the target image 4b and source image 4a before registration, after registration of one object by OF$^{dist}$, and after global registration by MIM
 $\| T - S \|_2$ $\| T - S_{1,OF^{dist}} \|_2$ $\| T - S_{MIM} \|_2$ 6654.44 5062.1 4121.95
 $\| T - S \|_2$ $\| T - S_{1,OF^{dist}} \|_2$ $\| T - S_{MIM} \|_2$ 6654.44 5062.1 4121.95
MI target image and source images from Figure 8 before and after registration by OF$^{dist}$ and MIM and MIM$^{dist}$. The MI was computed only in the surroundings of the segmented objects
 $i$ $MI(T,S_i)$ $MI(T,S_{i,OF^{dist}})$ $MI(T,S_{i,MIM})$ $MI(T,S_{i,MIM^{dist}})$ 1 1.1556 1.2184 1.2273 1.2170 2 1.1012 1.2034 1.2052 1.1964
 $i$ $MI(T,S_i)$ $MI(T,S_{i,OF^{dist}})$ $MI(T,S_{i,MIM})$ $MI(T,S_{i,MIM^{dist}})$ 1 1.1556 1.2184 1.2273 1.2170 2 1.1012 1.2034 1.2052 1.1964
 [1] Dana Paquin, Doron Levy, Eduard Schreibmann, Lei Xing. Multiscale Image Registration. Mathematical Biosciences & Engineering, 2006, 3 (2) : 389-418. doi: 10.3934/mbe.2006.3.389 [2] Zhao Yi, Justin W. L. Wan. An inviscid model for nonrigid image registration. Inverse Problems & Imaging, 2011, 5 (1) : 263-284. doi: 10.3934/ipi.2011.5.263 [3] Fan Jia, Xue-Cheng Tai, Jun Liu. Nonlocal regularized CNN for image segmentation. Inverse Problems & Imaging, 2020, 14 (5) : 891-911. doi: 10.3934/ipi.2020041 [4] Dana Paquin, Doron Levy, Lei Xing. Hybrid multiscale landmark and deformable image registration. Mathematical Biosciences & Engineering, 2007, 4 (4) : 711-737. doi: 10.3934/mbe.2007.4.711 [5] Christiane Pöschl, Jan Modersitzki, Otmar Scherzer. A variational setting for volume constrained image registration. Inverse Problems & Imaging, 2010, 4 (3) : 505-522. doi: 10.3934/ipi.2010.4.505 [6] Ye Yuan, Yan Ren, Xiaodong Liu, Jing Wang. Approach to image segmentation based on interval neutrosophic set. Numerical Algebra, Control & Optimization, 2020, 10 (1) : 1-11. doi: 10.3934/naco.2019028 [7] Dominique Zosso, Jing An, James Stevick, Nicholas Takaki, Morgan Weiss, Liane S. Slaughter, Huan H. Cao, Paul S. Weiss, Andrea L. Bertozzi. Image segmentation with dynamic artifacts detection and bias correction. Inverse Problems & Imaging, 2017, 11 (3) : 577-600. doi: 10.3934/ipi.2017027 [8] Matthew S. Keegan, Berta Sandberg, Tony F. Chan. A multiphase logic framework for multichannel image segmentation. Inverse Problems & Imaging, 2012, 6 (1) : 95-110. doi: 10.3934/ipi.2012.6.95 [9] Marcus Wagner. A direct method for the solution of an optimal control problem arising from image registration. Numerical Algebra, Control & Optimization, 2012, 2 (3) : 487-510. doi: 10.3934/naco.2012.2.487 [10] Angel Angelov, Marcus Wagner. Multimodal image registration by elastic matching of edge sketches via optimal control. Journal of Industrial & Management Optimization, 2014, 10 (2) : 567-590. doi: 10.3934/jimo.2014.10.567 [11] Yangang Chen, Justin W. L. Wan. Numerical method for image registration model based on optimal mass transport. Inverse Problems & Imaging, 2018, 12 (2) : 401-432. doi: 10.3934/ipi.2018018 [12] Shi Yan, Jun Liu, Haiyang Huang, Xue-Cheng Tai. A dual EM algorithm for TV regularized Gaussian mixture model in image segmentation. Inverse Problems & Imaging, 2019, 13 (3) : 653-677. doi: 10.3934/ipi.2019030 [13] Jianping Zhang, Ke Chen, Bo Yu, Derek A. Gould. A local information based variational model for selective image segmentation. Inverse Problems & Imaging, 2014, 8 (1) : 293-320. doi: 10.3934/ipi.2014.8.293 [14] Lu Tan, Ling Li, Senjian An, Zhenkuan Pan. Nonlinear diffusion based image segmentation using two fast algorithms. Mathematical Foundations of Computing, 2019, 2 (2) : 149-168. doi: 10.3934/mfc.2019011 [15] Ruiliang Zhang, Xavier Bresson, Tony F. Chan, Xue-Cheng Tai. Four color theorem and convex relaxation for image segmentation with any number of regions. Inverse Problems & Imaging, 2013, 7 (3) : 1099-1113. doi: 10.3934/ipi.2013.7.1099 [16] Balázs Kósa, Karol Mikula, Markjoe Olunna Uba, Antonia Weberling, Neophytos Christodoulou, Magdalena Zernicka-Goetz. 3D image segmentation supported by a point cloud. Discrete & Continuous Dynamical Systems - S, 2021, 14 (3) : 971-985. doi: 10.3934/dcdss.2020351 [17] Jie Huang, Xiaoping Yang, Yunmei Chen. A fast algorithm for global minimization of maximum likelihood based on ultrasound image segmentation. Inverse Problems & Imaging, 2011, 5 (3) : 645-657. doi: 10.3934/ipi.2011.5.645 [18] Liam Burrows, Weihong Guo, Ke Chen, Francesco Torella. Reproducible kernel Hilbert space based global and local image segmentation. Inverse Problems & Imaging, 2021, 15 (1) : 1-25. doi: 10.3934/ipi.2020048 [19] Huan Han. A variational model with fractional-order regularization term arising in registration of diffusion tensor image. Inverse Problems & Imaging, 2018, 12 (6) : 1263-1291. doi: 10.3934/ipi.2018053 [20] Yunmei Chen, Jiangli Shi, Murali Rao, Jin-Seop Lee. Deformable multi-modal image registration by maximizing Rényi's statistical dependence measure. Inverse Problems & Imaging, 2015, 9 (1) : 79-103. doi: 10.3934/ipi.2015.9.79

2020 Impact Factor: 2.425