Registration Evaluation Project Research Articles

MRI and CT Fusion in Stereotactic Electroencephalography (SEEG).

Epilepsy is a neurological disorder characterized by spontaneous recurrent seizures. While 20% to 30% of epilepsy cases are untreatable with Anti-Epileptic Drugs, some of these cases can be addressed through surgical intervention. The success of such interventions greatly depends on accurately locating the epileptogenic tissue, a task achieved using diagnostic techniques like Stereotactic Electroencephalography (SEEG). SEEG utilizes multi-modal fusion to aid in electrode localization, using pre-surgical resonance and post-surgical computer tomography images as inputs. To ensure the absence of artifacts or misregistrations in the resultant images, a fusion method that accounts for electrode presence is required. We proposed an image fusion method in SEEG that incorporates electrode segmentation from computed tomography as a sampling mask during registration to address the fusion problem in SEEG. The method was validated using eight image pairs from the Retrospective Image Registration Evaluation Project (RIRE). After establishing a reference registration for the MRI and identifying eight points, we assessed the method's efficacy by comparing the Euclidean distances between these reference points and those derived using registration with a sampling mask. The results showed that the proposed method yielded a similar average error to the registration without a sampling mask, but reduced the dispersion of the error, with a standard deviation of 0.86 when a mask was used and 5.25 when no mask was used.

Variability in commercially available deformable image registration: A multi‐institution analysis using virtual head and neck phantoms

PurposeThe purpose of this study was to evaluate the performance of three common deformable image registration (DIR) packages across algorithms and institutions.Methods and MaterialsThe Deformable Image Registration Evaluation Project (DIREP) provides ten virtual phantoms derived from computed tomography (CT) datasets of head‐and‐neck cancer patients over a single treatment course. Using the DIREP phantoms, DIR results from 35 institutions were submitted using either Velocity, MIM, or Eclipse. Submitted deformation vector fields (DVFs) were compared to ground‐truth DVFs to calculate target registration error (TRE) for six regions of interest (ROIs). Statistical analysis was performed to determine the variability between each DIR software package and the variability of users within each algorithm.ResultsOverall mean TRE was 2.04 ± 0.35 mm for Velocity, 1.10 ± 0.29 mm for MIM, and 2.35 ± 0.15 mm for Eclipse. The MIM mean TRE was significantly different than both Velocity and Eclipse for all ROIs. Velocity and Eclipse mean TREs were not significantly different except for when evaluating the registration of the cord or mandible. Significant differences between institutions were found for the MIM and Velocity platforms. However, these differences could be explained by variations in Velocity DIR parameters and MIM software versions.ConclusionsAverage TRE was shown to be <3 mm for all three software platforms. However, maximum errors could be larger than 2 cm indicating that care should be exercised when using DIR. While MIM performed statistically better than the other packages, all evaluated algorithms had an average TRE better than the largest voxel dimension. For the phantoms studied here, significant differences between algorithm users were minimal suggesting that the algorithm used may have more impact on DIR accuracy than the particular registration technique employed. A significant difference in TRE was discovered between MIM versions showing that DIR QA should be performed after software upgrades as recommended by TG‐132.

Band-Limited Stokes Large Deformation Diffeomorphic Metric Mapping.

The class of registration methods proposed in the framework of Stokes large deformation diffeomorphic metric mapping (LDDMM) is a particularly interesting family of physically meaningful diffeomorphic registration methods. Stokes-LDDMM methods are formulated as constrained variational problems, where the different physical models are imposed using the associated partial differential equations as hard constraints. The most significant limitation of Stokes-LDDMM framework is its huge computational complexity. The objective of this paper is to promote the use of Stokes-LDDMM in computational anatomy applications with an efficient approximation of the original variational problem. Thus, we propose a novel method for efficient Stokes-LDDMM diffeomorphic registration. Our method poses the constrained variational problem in the space of band-limited vector fields and it is implemented in the GPU. The performance of band-limited Stokes-LDDMM has been compared and evaluated with original Stokes-LDDMM, EPDiff-LDDMM, and band-limited EPDiff-LDDMM. The evaluation has been conducted in 3-D with the nonrigid image registration evaluation project database. Since the update equation in Stokes-LDDMM involves the action of low-pass filters, the computational complexity has been greatly alleviated with a modest accuracy lose. We have obtained a competitive performance for some method configurations. Overall, our proposed method may make feasible the extensive use of novel physically meaningful Stokes-LDDMM methods in different computational anatomy applications. In addition, our results reinforce the usefulness of band-limited vector fields in diffeomorphic registration methods involving the action of low-pass filters in the optimization, even in algorithmically challenging environments such as Stokes-LDDMM.

Potential of metric homotopy between intensity and geometry information for multi-modal 3D registration

This paper focuses on a novel strategy increasing robustness with respect to local optima when using Mutual Information (MI) in multi-modal image registration. This is realized by integrating additional geometry information in the cost function. Hereby, the main innovation is a generalization of multi-metric registration approaches by means of a metric homotopy. Particularly we realize a method which automatically determines the parameters of the metric homotopy. To construct the cost function independent of the choice of the optimizer, the weighting is defined as a function of one of the metrics instead of optimizer steps. In addition, a differentiable cost function is developed. In comparison to the commonly used technique to process an intensity based registration on different resolutions, the proposed method is three times faster with unchanged accuracy. It is also shown that in the presence of large landmark errors the proposed method outperforms an approach in accuracy in which both similarity functionals are applied one after the other.The method is evaluated on 3D multi-modal human brain data sets from the Retrospective Image Registration Evaluation Project (RIRE). The evaluation is performed using the evaluation website of the RIRE project to make the registration results of the proposed method easily comparable to other methods. Therefore, the presented results are also available online on the RIRE project page.

Erratum: Evaluation of the tool "Reg Refine" for user-guided deformable image registration.

"Reg Refine" is a tool available in the MIM Maestro v6.4.5 platform (www.mim-software.com) that allows the user to actively participate in the deformable image registration process. The purpose of this work was to evaluate the efficacy of this tool and investigate strategies for how to apply it effectively. This was done by performing DIR on two publicly available ground-truth models, the Pixel-based Breathing Thorax Model (POPI) for lung, and the Deformable Image Registration Evaluation Project (DIREP) for head and neck. Image noise matched in both magnitude and texture to clinical CBCT scans was also added to each model to simulate the use case of CBCT-CT alignment. For lung, the results showed Reg Refine effective at improving registration accuracy when controlled by an expert user within the context of large lung deformation. CBCT noise was also shown to have no effect on DIR performance while using the MIM algorithm for this site. For head and neck, the results showed CBCT noise to have a large effect on the accuracy of registration, specifically for low-contrast structures such as the brain-stem and parotid glands. In these cases, the Reg Refine tool was able to improve the registration accuracy when controlled by an expert user. Several strategies for how to achieve these results have been outlined to assist other users and provide feedback for developers of similar tools.

Gauss–Newton inspired preconditioned optimization in large deformation diffeomorphic metric mapping

In this work, we propose a novel preconditioned optimization method in the paradigm of Large Deformation Diffeomorphic Metric Mapping (LDDMM). The preconditioned update scheme is formulated for the non-stationary and the stationary parameterizations of diffeomorphisms, yielding three different LDDMM methods. The preconditioning matrices are inspired in the Hessian approximation used in Gauss–Newton method. The derivatives are computed using Frechet differentials. Thus, optimization is performed in a Sobolev space, in contrast to optimization in L2 commonly used in non-rigid registration literature. The proposed LDDMM methods have been evaluated and compared with their respective implementations of gradient descent optimization. Evaluation has been performed using real and simulated images from the Non-rigid Image Registration Evaluation Project (NIREP). The experiments conducted in this work reported that our preconditioned LDDMM methods achieved a performance similar or superior to well-established-in-literature gradient descent non-stationary LDDMM in the great majority of cases. Moreover, preconditioned optimization showed a substantial reduction in the execution time with an affordable increase of the memory usage per iteration. Additional experiments reported that optimization using Frechet differentials should be preferable to optimization using L2 differentials.

Multi-Modal Image Registration Based on Gradient Orientations of Minimal Uncertainty

In this paper, we propose a new multi-scale technique for multi-modal image registration based on the alignment of selected gradient orientations of reduced uncertainty. We show how the registration robustness and accuracy can be improved by restricting the evaluation of gradient orientation alignment to locations where the uncertainty of fixed image gradient orientations is minimal, which we formally demonstrate correspond to locations of high gradient magnitude. We also embed a computationally efficient technique for estimating the gradient orientations of the transformed moving image (rather than resampling pixel intensities and recomputing image gradients). We have applied our method to different rigid multi-modal registration contexts. Our approach outperforms mutual information and other competing metrics in the context of rigid multi-modal brain registration, where we show sub-millimeter accuracy with cases obtained from the retrospective image registration evaluation project. Furthermore, our approach shows significant improvements over standard methods in the highly challenging clinical context of image guided neurosurgery, where we demonstrate misregistration of less than 2 mm with relation to expert selected landmarks for the registration of pre-operative brain magnetic resonance images to intra-operative ultrasound images.

MULTI-FEATURE MUTUAL INFORMATION IMAGE REGISTRATION

Nowadays, information-theoretic similarity measures, especially the mutual information and its derivatives, are one of the most frequently used measures of global intensity feature correspondence in image registration. Because the traditional mutual information similarity measure ignores the dependency of intensity values of neighboring image elements, registration based on mutual information is not robust in cases of low global intensity correspondence. Robustness can be improved by adding spatial information in the form of local intensity changes to the global intensity correspondence. This paper presents a novel method, by which intensities, together with spatial information, i.e., relations between neighboring image elements in the form of intensity gradients, are included in information-theoretic similarity measures. In contrast to a number of heuristic methods that include additional features into the generic mutual information measure, the proposed method strictly follows information theory under certain assumptions on feature probability distribution. The novel approach solves the problem of efficient estimation of multifeature mutual information from sparse high-dimensional feature space. The proposed measure was tested on magnetic resonance (MR) and computed tomography (CT) images. In addition, the measure was tested on positron emission tomography (PET) and MR images from the widely used Retrospective Image Registration Evaluation project image database. The results indicate that multi-feature mutual information, which combines image intensities and intensity gradients, is more robust than the standard single-feature intensity based mutual information, especially in cases of low global intensity correspondences, such as in PET/MR images or significant intensity inhomogeneity.

Implicit reference-based group-wise image registration and its application to structural and functional MRI

In this study, an implicit reference group-wise (IRG) registration with a small deformation, linear elastic model was used to jointly estimate correspondences between a set of MRI images. The performance of pair-wise and group-wise registration algorithms was evaluated for spatial normalization of structural and functional MRI data. Traditional spatial normalization is accomplished by group-to-reference (G2R) registration in which a group of images are registered pair-wise to a reference image. G2R registration is limited due to bias associated with selecting a reference image. In contrast, implicit reference group-wise (IRG) registration estimates correspondences between a group of images by jointly registering the images to an implicit reference corresponding to the group average. The implicit reference is estimated during IRG registration eliminating the bias associated with selecting a specific reference image. Registration performance was evaluated using segmented T1-weighted magnetic resonance images from the Nonrigid Image Registration Evaluation Project (NIREP), DTI and fMRI images. Implicit reference pair-wise (IRP) registration—a special case of IRG registration for two images—is shown to produce better relative overlap than IRG for pair-wise registration using the same small deformation, linear elastic registration model. However, IRP-G2R registration is shown to have significant transitivity error, i.e., significant inconsistencies between correspondences defined by different pair-wise transformations. In contrast, IRG registration produces consistent correspondence between images in a group at the cost of slightly reduced pair-wise RO accuracy compared to IRP-G2R. IRG spatial normalization of the fractional anisotropy (FA) maps of DTI is shown to have smaller FA variance compared with G2R methods using the same elastic registration model. Analyses of fMRI data sets with sensorimotor and visual tasks show that IRG registration, on average, increases the statistical detectability of brain activation compared to G2R registration.

Registration of medical images using an interpolated closest point transform: method and validation

Image registration is an important procedure for medical diagnosis. Since the large inter-site retrospective validation study led by Fitzpatrick at Vanderbilt University, voxel-based methods and more specifically mutual information-based registration methods (see for instance [IEEE Trans. Med. Imag. 22 (8) (2003) 986] for a review on these methods) have been regarded as the method of choice for rigid-body intra-subject registration problems. In this study we propose a method that is based on the Iterative Closest Point algorithm and a pre-computed closest point map obtained with a slight modification of the fast marching method proposed by Sethian. Pre-computing the closest point map speeds up the process because at each iteration point correspondence can be established by table lookup. We also show that because the closest point map is defined on a regular grid it introduces a registration error and we propose an interpolation scheme that addresses this issue. The method has been tested both on synthetic and real images, and registration results have been assessed quantitatively using the data set provided by the Retrospective Registration Evaluation Project. For these volumes, MR and CT head surfaces were extracted automatically using a level-set technique. Results show that on these data sets this registration method leads to accuracy numbers that are comparable to those obtained with voxel-based methods.

Registration of functional and anatomical MRI: Accuracy assessment and application in navigated neurosurgery

A procedure for acquisition, automated registration, and fusion of functional and anatomical magnetic resonance images is presented. Its accuracy is quantitatively assessed using a publicly available gold standard. A patient case is used to illustrate the technique's clinical usefulness in image-guided neurosurgery. Before and after functional MRI (fMRI) acquisition, additional anatomical images were acquired at spatial locations identical to those of the functional images (5-10 slices) for the purpose of voxel-based image registration. Registration accuracy of the anatomical volumes and high-resolution 3D MRI volumes (MP-RAGE imaging) was quantified using adapted data (8 patients) originating from the Vanderbilt Retrospective Registration Evaluation Project (NIH project 1 R01 NS33926-02). Selecting three subsets of slices from that data (5 slices/6 mm slice distance, 10 slices/3 mm distance, and 10 slices/6 mm distance), the small number of images available from fMRI acquisition was taken into account. Accuracies in registering these sparse data sets were then compared to the accuracy achieved using complete data. For clinical patient data (16 patients), fMRI images were fused with MP-RAGE images, thereby integrating anatomical images with information about the locations of functional areas. The resulting images were used for planning and navigation during tumor resections using an operating microscope (MKM, Zeiss). Quantitative analysis showed no loss of registration accuracy due to a reduced number of slices, regardless of whether 5 or 10 slices were used. For small-volume coverage in the anatomical images (thickness 24 mm), registration of one patient failed, and this could easily be identified by visual inspection. No failures were experienced when 54 mm was covered. In the clinical environment, all 16 interventions using fused fMRI and MRI data were successful. Automatic registration of functional and high-resolution anatomical MRI was found to be sufficiently accurate and reliable for use in stereotactic neurosurgery.

Registration of Functional and Anatomical MRI: Accuracy Assessment and Application in Navigated Neurosurgery

Objective: A procedure for acquisition, automated registration, and fusion of functional and anatomical magnetic resonance images is presented. Its accuracy is quantitatively assessed using a publicly available gold standard. A patient case is used to illustrate the technique's clinical usefulness in image-guided neurosurgery.Materials and Methods: Before and after functional MRI (fMRI) acquisition, additional anatomical images were acquired at spatial locations identical to those of the functional images (5–10 slices) for the purpose of voxel-based image registration. Registration accuracy of the anatomical volumes and high-resolution 3D MRI volumes (MP-RAGE imaging) was quantified using adapted data (8 patients) originating from the Vanderbilt Retrospective Registration Evaluation Project (NIH project 1 R01 NS33926–02). Selecting three subsets of slices from that data (5 slices/6 mm slice distance, 10 slices/3 mm distance, and 10 slices/6 mm distance), the small number of images available from fMRI acquisition was taken into account. Accuracies in registering these sparse data sets were then compared to the accuracy achieved using complete data. For clinical patient data (16 patients), fMRI images were fused with MP-RAGE images, thereby integrating anatomical images with information about the locations of functional areas. The resulting images were used for planning and navigation during tumor resections using an operating microscope (MKM, Zeiss).Results: Quantitative analysis showed no loss of registration accuracy due to a reduced number of slices, regardless of whether 5 or 10 slices were used. For small-volume coverage in the anatomical images (thickness 24 mm), registration of one patient failed, and this could easily be identified by visual inspection. No failures were experienced when 54 mm was covered. In the clinical environment, all 16 interventions using fused fMRI and MRI data were successful.