Learning-based deformable registration of MR brain images

This paper presents a learning-based method for deformable registration of magnetic resonance (MR) brain images. There are two novelties in the proposed registration method. First, a set of best-scale geometric features are selected for each point in the brain, in order to facilitate correspondence detection during the registration procedure. This is achieved by optimizing an energy function that requires each point to have its best-scale geometric features consistent over the corresponding points in the training samples, and at the same time distinctive from those of nearby points in the neighborhood. Second, the active points used to drive the brain registration are hierarchically selected during the registration procedure, based on their saliency and consistency measures. That is, the image points with salient and consistent features (across different individuals) are considered for the initial registration of two images, while other less salient and consistent points join the registration procedure later. By incorporating these two novel strategies into the framework of the HAMMER registration algorithm, the registration accuracy has been improved according to the results on simulated brain data, and also visible improvement is observed particularly in the cortical regions of real brain data.     

Automatic registration and alignment on a template of cardiacstress and rest reoriented SPECT images

Single photon emission computed tomography (SPECT) imaging with ²⁰¹Tl or ^99mTc agent is used to assess the location or the extent of myocardial infarction or ischemia. A method is proposed to decrease the effect of operator variability in the visual or quantitative interpretation of scintigraphic myocardial perfusion studies. To effect this, the patient's myocardial images (target cases) are registered automatically over a template image, utilizing a nonrigid transformation. The intermediate steps are: 1) Extraction of feature points in both stress and rest three-dimensional (3-D) images. The images are resampled in a polar geometry to detect edge points, which in turn are filtered by the use of a priori constraints. The remaining feature points are assumed to be points on the edges of the left ventricular myocardium. 2) Registration of stress and rest images with a global affine transformation. The matching method is an adaptation of the iterative closest point algorithm. 3) Registration and morphological matching of both stress and rest images on a template using a nonrigid local spline transformation following a global affine transformation. 4) Resampling of both stress and rest images in the geometry of the template. Optimization of the method was performed on a database of 40 pairs of stress and rest images selected to obtain a wide variation of images and abnormalities. Further testing was performed on 250 cases selected from the same database on the basis of the availability of angiographic results and patient stratification     

A comparative study of biomechanical simulators in deformable registration of brain tumor images

Simulating the brain tissue deformation caused by tumor growth has been found to aid the deformable registration of brain tumor images. In this paper, we evaluate the impact that different biomechanical simulators have on the accuracy of deformable registration. We use two alternative frameworks for biomechanical simulations of mass effect in 3-D magnetic resonance (MR) brain images. The first one is based on a finite-element model of nonlinear elasticity and unstructured meshes using the commercial software package ABAQUS. The second one employs incremental linear elasticity and regular grids in a fictitious domain method. In practice, biomechanical simulations via the second approach may be at least ten times faster. Landmarks error and visual examination of the coregistered images indicate that the two alternative frameworks for biomechanical simulations lead to comparable results of deformable registration. Thus, the computationally less expensive biomechanical simulator offers a practical alternative for registration purposes.     

Improved image quality and computation reduction in 4-D reconstruction of cardiac-gated SPECT images

Spatiotemporal reconstruction of cardiac-gated SPECT images permits us to obtain valuable information related to cardiac function. However, the task of reconstructing this four-dimensional (4-D) data set is computation intensive. Typically, these studies are reconstructed frame-by-frame: a nonoptimal approach because temporal correlations in the signal are not accounted for. In this work, we show that the compression and signal decorrelation properties of the Karhunen-Loève (KL) transform may be used to greatly simplify the spatiotemporal reconstruction problem. The gated projections are first KL transformed in the temporal direction. This results in a sequence of KL-transformed projection images for which the signal components are uncorrelated along the time axis. As a result, the 4-D reconstruction task is simplified to a series of three-dimensional (3-D) reconstructions in the KL domain. The reconstructed KL components are subsequently inverse KL transformed to obtain the entire spatiotemporal reconstruction set. Our simulation and clinical results indicate that KL processing provides image sequences that are less noisy than are conventional frame-by-frame reconstructions. Additionally, by discarding high-order KL components that are dominated by noise, we can achieve savings in computation time because fewer reconstructions are needed in comparison to conventional frame-by-frame reconstructions.     

ORBIT: a multiresolution framework for deformable registration of brain tumor images

A deformable registration method is proposed for registering a normal brain atlas with images of brain tumor patients. The registration is facilitated by first simulating the tumor mass effect in the normal atlas in order to create an atlas image that is as similar as possible to the patient's image. An optimization framework is used to optimize the location of tumor seed as well as other parameters of the tumor growth model, based on the pattern of deformation around the tumor region. In particular, the optimization is implemented in a multiresolution and hierarchical scheme, and it is accelerated by using a principal component analysis (PCA)-based model of tumor growth and mass effect, trained on a computationally more expensive biomechanical model. Validation on simulated and real images shows that the proposed registration framework, referred to as ORBIT (optimization of tumor parameters and registration of brain images with tumors), outperforms other available registration methods particularly for the regions close to the tumor, and it has the potential to assist in constructing statistical atlases from tumor-diseased brain images.      

Feature-based registration of retinal images

Registration of retinal images taken at different times frequently is required to measure changes caused by disease or to document retinal location of visual stimuli. Cross-correlation has been used previously for such registration, but it is computationally intensive. We have modified a faster algorithm, sequential similarity detection (SSD), to use only the portion of the template that contains retinal vessels. When compared to standard SSD and cross-correlation, this modification improves the reliability of detection for a variety of retinal imaging modalities. The improved reliability enables implementation of a two-stage registration strategy that further decreases the amount of computation and increases the speed of registration.     

A deformable registration method for automated morphometry of MRI brain images in neuropsychiatric research

Image registration methods play a crucial role in computational neuroanatomy. This paper mainly contributes to the field of image registration with the use of nonlinear spatial transformations. Particularly, problems connected to matching magnetic resonance imaging (MRI) brain image data obtained from various subjects and with various imaging conditions are solved here. Registration is driven by local forces derived from multimodal point similarity measures which are estimated with the use of joint intensity histogram and tissue probability maps. A spatial deformation model imitating principles of continuum mechanics is used. Five similarity measures are tested in an experiment with image data obtained from the Simulated Brain Database and a quantitative evaluation of the algorithm is presented. Results of application of the method in automated spatial detection of anatomical abnormalities in first-episode schizophrenia are presented.     

Automatic registration of CT and MR brain images using correlation of geometrical features

Describes an automated approach to register CT and MR brain images. Differential operators in scale space are applied to each type of image data, so as to produce feature images depicting "ridgeness". The resulting CT and MR feature images show similarities which can be used for matching. No segmentation is needed and the method is devoid of human interaction. The matching is accomplished by hierarchical correlation techniques. Results of 2-D and 3-D matching experiments are presented. The correlation function ensures an accurate match even if the scanned volumes to be matched do not completely overlap, or if some of the features in the images are not similar.     

Validation of nonrigid image registration using finite-element methods: application to breast MR images

This paper presents a novel method for validation of nonrigid medical image registration. This method is based on the simulation of physically plausible, biomechanical tissue deformations using finite-element methods. Applying a range of displacements to finite-element models of different patient anatomies generates model solutions which simulate gold standard deformations. From these solutions, deformed images are generated with a range of deformations typical of those likely to occur in vivo. The registration accuracy with respect to the finite-element simulations is quantified by co-registering the deformed images with the original images and comparing the recovered voxel displacements with the biomechanically simulated ones. The functionality of the validation method is demonstrated for a previously described nonrigid image registration technique based on free-form deformations using B-splines and normalized mutual information as a voxel similarity measure, with an application to contrast-enhanced magnetic resonance mammography image pairs. The exemplar nonrigid registration technique is shown to be of subvoxel accuracy on average for this particular application. The validation method presented here is an important step toward more generic simulations of biomechanically plausible tissue deformations and quantification of tissue motion recovery using nonrigid image registration. It will provide a basis for improving and comparing different nonrigid registration techniques for a diversity of medical applications, such as intrasubject tissue deformation or motion correction in the brain, liver or heart.     

Three-dimensional blind deconvolution of SPECT images

Thanks to its ability to yield functionally rather than anatomically-based information, the three-dimensional (3-D) SPECT imagery technique has become a great help in the diagnostic of cerebrovascular diseases. Nevertheless, due to the imaging process, the 3-D single photon emission computed tomography (SPECT) images are very blurred and, consequently, their interpretation by the clinician is often difficult and subjective. In order to improve the resolution of these 3-D images and then to facilitate their interpretation, we propose herein to extend a recent image blind deconvolution technique (called the nonnegativity support constraint-recursive inverse filtering deconvolution method) in order to improve both the spatial and the interslice resolution of SPECT volumes. This technique requires a preliminary step in order to find the support of the object to be restored. In this paper, we propose to solve this problem with an unsupervised 3-D Markovian segmentation technique. This method has been successfully tested on numerous real and simulated brain SPECT volumes, yielding very promising restoration results.     

Improved labeling of subcortical brain structures in atlas-based segmentation of magnetic resonance images

Precise labeling of subcortical structures plays a key role in functional neurosurgical applications. Labels from an atlas image are propagated to a patient image using atlas-based segmentation. Atlas-based segmentation is highly dependent on the registration framework used to guide the atlas label propagation. This paper focuses on atlas-based segmentation of subcortical brain structures and the effect of different registration methods on the generated subcortical labels. A single-step and three two-step registration methods appearing in the literature based on affine and deformable registration algorithms in the ANTS and FSL algorithms are considered. Experiments are carried out with two atlas databases of IBSR and LPBA40. Six segmentation metrics consisting of Dice overlap, relative volume error, false positive, false negative, surface distance, and spatial extent are used for evaluation. Segmentation results are reported individually and as averages for nine subcortical brain structures. Based on two statistical tests, the results are ranked. In general, among four different registration strategies investigated in this paper, a two-step registration consisting of an initial affine registration followed by a deformable registration applied to subcortical structures provides superior segmentation outcomes. This method can be used to provide an improved labeling of the subcortical brain structures in MRIs for different applications.     

Stochastic methods for joint registration, restoration, and interpolation of multiple undersampled images.

Using a stochastic framework, we propose two algorithms for the problem of obtaining a single high-resolution image from multiple noisy, blurred, and undersampled images. The first is based on a Bayesian formulation that is implemented via the expectation maximization algorithm. The second is based on a maximum a posteriori formulation. In both of our formulations, the registration, noise, and image statistics are treated as unknown parameters. These unknown parameters and the high-resolution image are estimated jointly based on the available observations. We present an efficient implementation of these algorithms in the frequency domain that allows their application to large images. Simulations are presented that test and compare the proposed algorithms.     

Fusion of multi-focus images with registration inaccuracies

Image fusion combines complementary information for several input images. To obtain useful information from two misaligned images, registration is required. A hybrid textural registration-based multi-focus image fusion scheme is proposed. The Gabor filtering with specific frequency and orientation is used to extract different texture features from the image. The resulting Gabor-filtered images are then aligned using existing affine transformation. The proposed registration scheme yields superior performance as compared to affine registration, as Gabor transform extracts all the features. The fusion is performed using undecimated dual-tree complex wavelet transform. The quantitative and qualitative analysis of the proposed scheme outperforms existing image fusion schemes.     

Nonlinear elastic registration of brain images with tumor pathologyusing a biomechanical model [MRI]

A biomechanical model of the brain is presented, using a finite-element formulation. Emphasis is given to the modeling of the soft-tissue deformations induced by the growth of tumors and its application to the registration of anatomical atlases, with images from patients presenting such pathologies. First, an estimate of the anatomy prior to the tumor growth is obtained through a simulated biomechanical contraction of the tumor region. Then a normal-to-normal atlas registration to this estimated pre-tumor anatomy is applied. Finally, the deformation from the tumor-growth model is applied to the resultant registered atlas, producing an atlas that has been deformed to fully register to the patient images. The process of tumor growth is simulated in a nonlinear optimization framework, which is driven by anatomical features such as boundaries of brain structures. The deformation of the surrounding tissue is estimated using a nonlinear elastic model of soft tissue under the boundary conditions imposed by the skull, ventricles, and the falx and tentorium. A preliminary two-dimensional (2-D) implementation is presented in this paper, and tested on both simulated and patient data. One of the long-term goals of this work is to use anatomical brain atlases to estimate the locations of important brain structures in the brain and to use these estimates in presurgical and radiosurgical planning systems.     

Block-coordinate Gauss-Newton optimization and constrained monotone regression for image registration in the presence of outlier objects.

In this paper, we propose the block-coordinate Gauss- Newton/regression method in order to conduct a correlation-based registration considering the intensity difference between images in the presence of outlier objects. In the proposed method, the parameters are decomposed into two blocks, one of which is for the spatial registration and the other for the intensity compensation. The two blocks are sequentially updated by the Gauss-Newton update and the polynomial regression, respectively. Because of the separated blocks, we can perform a joint optimization with low computational complexity and high implementation flexibility. For example, we apply separately appropriate scaling techniques to the parameter blocks for a stable and fast convergence of the algorithm. Furthermore, we apply the constrained monotone regression with a robust outlier detection scheme for the intensity compensation block. From numerical results, it is shown that the proposed algorithm more effectively performs a correlation-based registration considering the intensity difference alleviating the influence of the outlier objects compared to the traditional registration algorithms that perform the joint optimization.     

医学图像的非刚性配准技术研究

非刚性配准技术是医学图像领域的一个重要研究方向，对于图像的进一步分析起着非常重要的作用。目前也已经提出了很多种解决方法，但是算法的成熟性远低于刚性配准，在临床中几乎没有应用。现对这些方法做了一个回顾，并详细阐述了非刚性配准的分类、方法、评估及研究进展。     

Elastic registration of biological images using vector-spline regularization

We present an elastic registration algorithm for the alignment of biological images. Our method combines and extends some of the best techniques available in the context of medical imaging. We express the deformation field as a B-spline model, which allows us to deal with a rich variety of deformations. We solve the registration problem by minimizing a pixelwise mean-square distance measure between the target image and the warped source. The problem is further constrained by way of a vector-spline regularization which provides some control over two independent quantities that are intrinsic to the deformation: its divergence, and its curl. Our algorithm is also able to handle soft landmark constraints, which is particularly useful when parts of the images contain very little information or when its repartition is uneven. We provide an optimal analytical solution in the case when only landmarks and smoothness considerations are taken into account. We have applied our approach to perform the elastic registration of images such as electrophoretic gels and fly embryos. The validation of the results by experts has been favorable in all cases.