Estimation of slipping organ motion by registration with direction-dependent regularization

Schmidt-Richberg, Alexander; Werner, René; Handels, Heinz; Ehrhardt, Jan

doi:10.1016/j.media.2011.06.007

Cited by 88 publications

(79 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Attempts to address this issue have been based on direction-dependent or spatially varying regularisers (Schmidt-Richberg et al, 2012;Pace et al, 2013;Ruan et al, 2009;Heinrich et al, 2013;Papież et al, 2014), multiple B-spline trans- formations (Wu et al, 2008;Delmon et al, 2013), or introduction of penalty terms to reduce gaps and overlaps (Berendsen et al, 2014). In contrast to these methods, we propose to handle this problem in the transformation model, such that the discontinuities are implicitly incorporated into a single B-spline transformation, while retaining all the desirable properties of Bsplines.…”

Section: Discussionmentioning

confidence: 99%

“…This requirement is, however, shared by all the other B-spline based methods (Wu et al, 2008;Delmon et al, 2013;Berendsen et al, 2014) and most of the diffusion-based methods ( Schmidt-Richberg et al, 2012;Risser et al, 2013;Pace et al, 2013). This may be facilitated by using an automatic segmentation method (Nakagomi et al, 2013;Tomoshige et al, 2014;Rebouças Filho et al, 2017) as a previous step.…”

Section: Discussionmentioning

confidence: 99%

“…The DIR-lab dataset was adopted in most of the previous methods treating discontinuities. We computed the same measurements, TRE, gap and overlap volumes, as reported in all the B-spline based methods (Wu et al, 2008;Delmon et al, 2013;Berendsen et al, 2014) and some diffusion-based methods (Schmidt-Richberg et al, 2012;Pace et al, 2013;Papież et al, 2014).…”

Section: Comparison With Previous Methodsmentioning

confidence: 99%

“…In most diffusionbased methods, discontinuities were explicitly incorporated in the regularization schemes. Direction-dependent regularizers were employed by decomposing the deformation field into normal and tangential directions at the discontinuity interface and smoothing was only applied in the tangential components but not across the boundary (Schmidt-Richberg et al, 2012;Pace et al, 2013). Locally adaptive regularizers were adopted with discontinuity preserving properties (Ruan et al, 2009;Papież et al, 2014).…”

Section: Previous Workmentioning

confidence: 99%

“…We computed the Euclidean distance between the landmarks in the moving image and those in the target image, displaced by the deformation field. The mean and standard deviation of the distance of all the landmarks in each case was reported as the TRE Target Registration Error (TRE) (Schmidt-Richberg et al, 2012;Pace et al, 2013;Papież et al, 2014;Wu et al, 2008;Delmon et al, 2013;Berendsen et al, 2014).…”

Section: Metrics For Evaluationmentioning

confidence: 99%

See 4 more Smart Citations

Multiresolution eXtended Free-Form Deformations (XFFD) for non-rigid registration with discontinuous transforms

Hua

Pozo

Taylor

et al. 2017

Medical Image Analysis

View full text Add to dashboard Cite

Image registration is an essential technique to obtain point correspondences between anatomical structures from different images. Conventional nonrigid registration methods assume a continuous and smooth deformation field throughout the image. However, the deformation field at the interface of different organs is not necessarily continuous, since the organs may slide over or separate from each other. Therefore, imposing continuity and smoothness ubiquitously would lead to artifacts and increased errors near the discontinuity interface.In computational mechanics, the eXtended Finite Element Method (XFEM) was introduced to handle discontinuities without using computational meshes that conform to the discontinuity geometry. Instead, the interpolation bases themselves were enriched with discontinuous functional terms. Borrowing this concept, we propose a multiresolution eXtented Free-Form Deformation (XFFD) framework that seamlessly integrates within and extends the standard Free-Form Deformation (FFD) approach. Discontinuities are incor- Preprint submitted to ElsevierOctober 18, 2016 porated by enriching the B-spline basis functions coupled with extra degrees of freedom, which are only introduced near the discontinuity interface. In contrast with most previous methods, restricted to sliding motion, no ad hoc penalties or constraints are introduced to reduce gaps and overlaps. This allows XFFD to describe more general discontinuous motions. In addition,we integrate XFFD into a rigorously formulated multiresolution framework by introducing an exact parameter upsampling method.The proposed method has been evaluated in two publicly available datasets:4D pulmonary CT images from the DIR-Lab dataset and 4D CT liver datasets.The XFFD achieved a Target Registration Error (TRE) of 1.17 ± 0.85 mm in the DIR-lab dataset and 1.94 ± 1.01 mm in the liver dataset, which significantly improves on the performance of the state-of-the-art methods handling discontinuities.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Comparison With Previous Methodsmentioning

confidence: 99%

Section: Previous Workmentioning

confidence: 99%

Section: Metrics For Evaluationmentioning

confidence: 99%

See 3 more Smart Citations

Multiresolution eXtended Free-Form Deformations (XFFD) for non-rigid registration with discontinuous transforms

Hua

Pozo

Taylor

et al. 2017

Medical Image Analysis

View full text Add to dashboard Cite

show abstract

An adaptive motion regularization technique to support sliding motion in deformable image registration

Shi

et al. 2018

Medical Physics

View full text Add to dashboard Cite

show abstract

LungRegNet: An unsupervised deformable image registration method for 4D‐CT lung

Lei

Wang

et al. 2020

Medical Physics

View full text Add to dashboard Cite

Purpose: To develop an accurate and fast deformable image registration (DIR) method for fourdimensional computed tomography (4D-CT) lung images. Deep learning-based methods have the potential to quickly predict the deformation vector field (DVF) in a few forward predictions. We have developed an unsupervised deep learning method for 4D-CT lung DIR with excellent performances in terms of registration accuracies, robustness, and computational speed. Methods: A fast and accurate 4D-CT lung DIR method, namely LungRegNet, was proposed using deep learning. LungRegNet consists of two subnetworks which are CoarseNet and FineNet. As the name suggests, CoarseNet predicts large lung motion on a coarse scale image while FineNet predicts local lung motion on a fine scale image. Both the CoarseNet and FineNet include a generator and a discriminator. The generator was trained to directly predict the DVF to deform the moving image. The discriminator was trained to distinguish the deformed images from the original images. Coarse-Net was first trained to deform the moving images. The deformed images were then used by the Fine-Net for FineNet training. To increase the registration accuracy of the LungRegNet, we generated vessel-enhanced images by generating pulmonary vasculature probability maps prior to the network prediction. Results: We performed fivefold cross validation on ten 4D-CT datasets from our department. To compare with other methods, we also tested our method using separate 10 DIRLAB datasets that provide 300 manual landmark pairs per case for target registration error (TRE) calculation. Our results suggested that LungRegNet has achieved better registration accuracy in terms of TRE than other deep learning-based methods available in the literature on DIRLAB datasets. Compared to conventional DIR methods, LungRegNet could generate comparable registration accuracy with TRE smaller than 2 mm. The integration of both the discriminator and pulmonary vessel enhancements into the network was crucial to obtain high registration accuracy for 4D-CT lung DIR. The mean and standard deviation of TRE were 1.00 AE 0.53 mm and 1.59 AE 1.58 mm on our datasets and DIRLAB datasets respectively. Conclusions: An unsupervised deep learning-based method has been developed to rapidly and accurately register 4D-CT lung images. LungRegNet has outperformed its deep-learning-based peers and achieved excellent registration accuracy in terms of TRE.

show abstract

Estimation of slipping organ motion by registration with direction-dependent regularization

Cited by 88 publications

References 36 publications

Multiresolution eXtended Free-Form Deformations (XFFD) for non-rigid registration with discontinuous transforms

Multiresolution eXtended Free-Form Deformations (XFFD) for non-rigid registration with discontinuous transforms

An adaptive motion regularization technique to support sliding motion in deformable image registration

LungRegNet: An unsupervised deformable image registration method for 4D‐CT lung

Contact Info

Product

Resources

About