Joint image segmentation and registration based on a dynamic level set approach using truncated hierarchical B-splines

Pawar, Aishwarya; Zhang, Yongjie; Anitescu, Cosmin; Rabczuk, Timon

doi:10.1016/j.camwa.2019.04.026

Cited by 11 publications

(12 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The spline-based segmentation technique reviewed above has been demonstrated to yield computational domains that are very well suited for isogeometric analysis (see, e.g., [36,58,72]). However, although the smoothing characteristic of the technique is frequently beneficial, it may lead to the occurrence of topological anomalies when the features of the object to be described are not significantly larger in size than the voxels (i.e., the Nyquist criterion is not satisfied [19]).…”

Section: The Occurrence Of Topological Anomaliesmentioning

confidence: 99%

“…[19] associated with the order of the spline level set construction, can trigger undesirable topology alterations (e.g., closure of a channel, (dis)connection of structures). The occurrence of topological anomalies associated with smoothing operations is well understood in the field of image segmentation (e.g., [70][71][72][73]). Various enhanced image-segmentation techniques have been proposed in order to ameliorate such problems, like homology-based preservation techniques [74][75][76][77][78][79][80], topology-derivative-based techniques [81][82][83][84][85], and adaptive refinement techniques [72,[86][87][88][89][90][91].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Topology-preserving scan-based immersed isogeometric analysis

Divi

Verhoosel

Auricchio

et al. 2022

Computer Methods in Applied Mechanics and Engineering

View full text Add to dashboard Cite

Section: The Occurrence Of Topological Anomaliesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Topology-preserving scan-based immersed isogeometric analysis

Divi

Verhoosel

Auricchio

et al. 2022

Computer Methods in Applied Mechanics and Engineering

View full text Add to dashboard Cite

“…Registration based on free-form deformation (FFD) using B-splines has emerged recently as a powerful tool in image analysis due to the smoothness and local control of B-spline basis functions [3][4][5][6][7]. In our earlier work [8], we developed a registration framework based on Bsplines, which allowed smooth, diffeomorphic and large deformations of 3D space. One of our main contributions in that earlier work was local refinement using truncated hierarchical B-splines (THB-splines) to maximize computational efficiency.…”

Section: Introductionmentioning

confidence: 99%

PDE-constrained shape registration to characterize biological growth and morphogenesis from imaging data

Pawar¹,

Li²,

Gosain³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

We propose a PDE-constrained shape registration algorithm that captures the deformation and growth of biological tissue from imaging data. Shape registration is the process of evaluating optimum alignment between pairs of geometries through a spatial transformation function. We start from our previously reported work, which uses 3D tensor product B-spline basis functions to interpolate 3D space. Here, the movement of the B-spline control points, composed with an implicit function describing the shape of the tissue, yields the total deformation gradient field. The deformation gradient is then split into growth and elastic contributions. The growth tensor captures addition of mass, i.e. growth, and evolves according to a constitutive equation which is usually a function of the elastic deformation. Stress is generated in the material due to the elastic component of the deformation alone. The result of the registration is obtained by minimizing a total energy functional which includes: a distance measure reflecting similarity between the shapes, and the total elastic energy accounting for the growth of the tissue.

show abstract

“…In [4], the authors formulated the joint problem as a minimization of a functional that integrates a nonlinear elastic registration with a geodesic active contours which is introduced together with a weighted total variation term to segment the deforming template image. Pawar et al [30] presented a joint approach using bidirectional composition based level set formulation, in which the implicit level set function defining the segmentation contour and the displacement field for registration are both defined using B-splines. Swier-czynski et al [37] proposed an algorithm based on a level-set formulation, which merges a classic Chan-Vese segmentation with an active dense displacement field estimation.…”

mentioning

confidence: 99%

“…However, these variational methods have some drawbacks. First, all the above methods [44,15,4,30,37,10] are designed for segmentation of two classes, and the extension to multi classes is not straightforward. Second, some of the joint models [44,4,10] are still intensity based methods, as they only consider the pixel value information from two images, without using any atlas as a shape prior.…”

mentioning

confidence: 99%

Image Segmentation with Adaptive Spatial Priors from Joint Registration

Li¹,

Guo²,

Li³

et al. 2022

Preprint

View full text Add to dashboard Cite

Image segmentation is a crucial but challenging task that has many applications. In medical imaging for instance, intensity inhomogeneity and noise are common. In thigh muscle images, different muscles are closed packed together and there are often no clear boundaries between them. Intensity based segmentation models cannot separate one muscle from another. To solve such problems, in this work we present a segmentation model with adaptive spatial priors from joint registration. This model combines segmentation and registration in a unified framework to leverage their positive mutual influence. The segmentation is based on a modified Gaussian mixture model (GMM), which integrates intensity inhomogeneity and spacial smoothness. The registration plays the role of providing a shape prior. We adopt a modified sum of squared difference (SSD) fidelity term and Tikhonov regularity term for registration, and also utilize Gaussian pyramid and parametric method for robustness. The connection between segmentation and registration is guaranteed by the cross entropy metric that aims to make the segmentation map (from segmentation) and deformed atlas (from registration) as similar as possible. This joint framework is implemented within a constraint optimization framework, which leads to an efficient algorithm. We evaluate our proposed model on synthetic and thigh muscle MR images. Numerical results show the improvement as compared to segmentation and registration performed separately and other joint models.

show abstract

Joint image segmentation and registration based on a dynamic level set approach using truncated hierarchical B-splines

Cited by 11 publications

References 35 publications

Topology-preserving scan-based immersed isogeometric analysis

Topology-preserving scan-based immersed isogeometric analysis

PDE-constrained shape registration to characterize biological growth and morphogenesis from imaging data

Image Segmentation with Adaptive Spatial Priors from Joint Registration

Contact Info

Product

Resources

About