A Scale-Space Approach to Landmark Constrained Image Registration

Haber, Eldad; Heldmann, Stefan; Modersitzki, Jan

doi:10.1007/978-3-642-02256-2_51

Cited by 11 publications

(8 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The contribution of this paper is twofold. First, a new pipeline for lung CT registration is introduced comprising of an automatic landmark detection algorithm [12] and a hybrid landmark-and intensitybased technique [21,22]. The second contribution is a thorough quantitative evaluation and statistical testing of the underlying hypothesis that the combination of these two building blocks is more accurate than each approach solely used.…”

Section: Discussionmentioning

confidence: 99%

“…For integration of the automatically detected landmark correspondences into the registration scheme MLIR described in the previous section, the so-called CoLD method is exploited. CoLD was introduced in [21] and later adjusted to the Discretizethen-Optimize scheme [22] used in this work. The fundamental idea of CoLD is to restrict the MLIR scheme to transformations that map the landmarks onto their corresponding locations in the other image.…”

Section: Combining Landmarks and Distance Measures (Cold)mentioning

confidence: 99%

See 1 more Smart Citation

Lung Registration Using Automatically Detected Landmarks

Polzin

Rühaak²,

Werner³

et al. 2014

Methods Inf Med

View full text Add to dashboard Cite

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Combining Landmarks and Distance Measures (Cold)mentioning

confidence: 99%

Lung Registration Using Automatically Detected Landmarks

Polzin

Rühaak²,

Werner³

et al. 2014

Methods Inf Med

View full text Add to dashboard Cite

show abstract

“…Scale spaces are an example for such a setting (Aldroubi & Grö chening, 2001;Florack, Romeny, Koenderink, & Viergever, 1992;Haber & Modersitzki, 2005, 2006cHaber et al, 2009b;Lindeberg, 1994;Salden, Ter Haar Romeny, & Viergever, 1998). They provide a great potential for a process that starts with the important coarse features, adds more and more fine details and structure, and ends with the original representation (see Modersitzki, 2009 for details).…”

Section: Data Modelmentioning

confidence: 99%

Nonlinear Registration Via Displacement Fields

2015

View full text Add to dashboard Cite

“…Specifically, the landmark kernel is supported by our framework. The diffeomorphic regularization is also approximated, as shown in Equation (14). Moreover, we showed a multiscale approach that brings the landmark mean together with the image-based optimization on different resolution levels.…”

Section: The Algorithmmentioning

confidence: 99%

“…Different hybrid methods, which combine landmarks and image features have been proposed in literature. For example in [14,16,26], the landmarks are treated as additional constraints. The methods require a perfect interpolation of the landmarks resulting in numerical problems during optimization.…”

Section: Introductionmentioning

confidence: 99%

Variational Image Registration Using Inhomogeneous Regularization

et al. 2014

View full text Add to dashboard Cite

We present a generalization of the convolutionbased variational image registration approach, in which different regularizers can be implemented by conveniently exchanging the convolution kernel, even if it is nonseparable or nonstationary. Nonseparable kernels pose a challenge because they cannot be efficiently implemented by separate 1D convolutions. We propose to use a low-rank tensor decomposition to efficiently approximate nonseparable convolution. Nonstationary kernels pose an even greater challenge because the convolution kernel depends on, and needs to be evaluated for, every point in the image. We propose to pre-compute the local kernels and efficiently store them in memory using the Tucker tensor decomposition model. In our experiments we use the nonseparable exponential kernel and a nonstationary landmark kernel. The exponential kernel replicates desirable properties of elastic image registration, while the landmark kernel incorporates local prior knowledge about corresponding points in the images. We examine the trade-off between the computational resources needed and the approximation accuracy of the tensor decomposition methods. Furthermore, we obtain very smooth displacement fields even in the presence of large landmark displacements.

show abstract

A Scale-Space Approach to Landmark Constrained Image Registration

Cited by 11 publications

References 19 publications

Lung Registration Using Automatically Detected Landmarks

Lung Registration Using Automatically Detected Landmarks

Nonlinear Registration Via Displacement Fields

Variational Image Registration Using Inhomogeneous Regularization

Contact Info

Product

Resources

About