Multimodal Non-Rigid Registration Methods Based on Demons Models and Local Uncertainty Quantification Used in 3D Brain Images (original) (raw)
Related papers
IET Image Processing, 2014
Non-rigid registration (NRR) of multimodal images is a common task in diagnostic and therapeutic procedures based on medical imaging, which at the same time imposes remarkable technical challenges. This work presents a novel non-rigid multimodal registration method that relies on three basic steps: first, an initial approximation of the deformation field is obtained by a parametric registration technique based on particle filtering; second, an intensity mapping based on local variability measures (LVM) is applied over the two images in order to overcome the multimodal restriction between them; and third, an optical flow method is used in an iterative way to find the remaining displacements of the deformation field. Hence the new methodology offers a solution for multimodal NRR by a quadratic optimization over a convex surface, which allows independent motion of each pixel, in contrast to methods that parameterize the deformation space. To evaluate the proposed method, a set of MR/CT clinical studies (pre-and post-radiotherapy treatment) of three patients with cerebral tumor deformations of the brain structures was employed. The resulting registration was evaluated both qualitatively and quantitatively by standard indices of correspondence over anatomical structures of interest in radiotherapy (brain, tumor and cerebral ventricles). These results showed that one of the proposed LVM (entropy) offers a superior performance in estimating the non-rigid deformation field.
Non-rigid multimodal image registration based on local variability measures and optical flow
2012 Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 2012
In this paper, we present a novel methodology for multimodal non-rigid medical image registration. The proposed approach is based on combining an optical flow technique with a pixel intensity transformation by using a local variability measure, such as statistical variance or Shannon entropy. The methodology is basically composed by three steps: first, we approximate the global deformation using a rigid registration based on a global optimization technique, called particle filtering; second, we transform both target and source images into a new intensity space where they can be compared; and third, we obtain the optical flow between them by using the Horn and Shuck algorithm in an iterative scales-space framework. After these steps, the non-rigid registration is made up by adding the resulting vector fields, computed by the rigid registration, and the optical flow. The proposed algorithm was tested using a synthetic intensity mapping and non-rigid deformation of MRI images. Preliminary results show that the methodology seems to be a good alternative for non-rigid multimodal registration, obtaining an average error of less than two pixels in the estimation of the deformation vector field.
IX International Seminar on Medical Information Processing and Analysis, 2013
In this work, we present a novel fully automated elastic registration method for magnetic resonance (MR) images with mismatched intensities, which combines a novel mapping based on an intensity uncertainty quantification in a local region, with a fluid-like registration technique. The proposed methodology can be summarized in two global steps: first, a mapping over the target and source images is applied, which provides information about the intensities uncertainty of the pixels in a neighborhood; and second, a monomodal non-rigid registration is achieved between the transformed images based on fluid-models: demons, diffeomorphic-demons, and a variation of the classical optical-flow. To evaluate the algorithm, a set composed by 12 multiparametric MR images of the head (T1, T2 and proton density) were taken from a brain model, and these images were modified by a set of controlled elastic deformations (based on splines), in order to generate ground-truths to be registered with the proposed technique. The evaluation results showed an average error of less than 1.3 mm by combining the local uncertainty quantification with the diffeomorphic-demons technique, which also ensures to obtain only feasible physical deformations. These results suggest that the proposed methodology could be considered as a good option for fully automated non-rigid registrations between images with mismatched intensities on medical applications.
Elastic Registration Based on Particle Filter in Radiotherapy Images with Brain Deformations
This paper presents the evaluation of the accuracy of an elastic registration algorithm, based on the particle filter and an optical flow process. The algorithm is applied in brain CT and MRI simulated image datasets, and MRI images from a real clinical radiotherapy case. To validate registration accuracy, standard indices for registration accuracy assessment were calculated: the dice similarity coefficient (DICE), the average symmetric distance (ASD) and the maximal distance between pixels (Dmax). The results showed that this registration process has good accuracy, both qualitatively and quantitatively, suggesting that this method may be considered as a good new option for radiotherapy applications like patient's follow up treatment.
Procedia Technology, 2013
In this paper, we present a new methodology for multimodal non-rigid medical image registration. The proposed approach is based on combining a rigid registration achieved by a global optimization method, and a multimodal optical flow technique based on the conditional statistics of the joint intensity distribution (CS-JID) of the images to register. The methodology is essentially composed of two steps: first, the global deformation is approximated by using a rigid registration based on particle filtering; second, the optical flow is applied iteratively between the target and sequentially registered source image, by optimizing a new energy function that penalizes the difference between the intensities in one image with respect to the mean of the conditional intensity distribution of the other image, weighted by the conditional variance. After these steps, the non-rigid registration is made up by adding the resulting vector fields, computed by the rigid registration, and the sequential optical flow. The proposed algorithm was tested with three pairs of computed tomography (CT) and magnetic resonance (MR) images, aligned in the acquisition process, and subsequently warped with a synthetic non-rigid deformation. Preliminary results show that the methodology presents a good alternative for non-rigid multimodal registration, obtaining an average error of less than one pixel in the estimation of the deformation vector field in the majority of the cases.
Nonrigid multimodal medical image registration using features extracted from the monogenic signal
In this paper, we present a new methodology for multimodal non-rigid medical image registration. The proposed approach is based on combining a rigid registration achieved by a global optimization method, and a multimodal optical flow technique based on the conditional statistics of the joint intensity distribution (CS-JID) of the images to register. The methodology is essentially composed of two steps: first, the global deformation is approximated by using a rigid registration based on particle filtering; second, the optical flow is applied iteratively between the target and sequentially registered source image, by optimizing a new energy function that penalizes the difference between the intensities in one image with respect to the mean of the conditional intensity distribution of the other image, weighted by the conditional variance. After these steps, the non-rigid registration is made up by adding the resulting vector fields, computed by the rigid registration, and the sequential optical flow. The proposed algorithm was tested with three pairs of computed tomography (CT) and magnetic resonance (MR) images, aligned in the acquisition process, and subsequently warped with a synthetic non-rigid deformation. Preliminary results show that the methodology presents a good alternative for non-rigid multimodal registration, obtaining an average error of less than one pixel in the estimation of the deformation vector field in the majority of the cases.
A hybrid multimodal non-rigid registration of MR images based on diffeomorphic demons
2010 Annual International Conference of the IEEE Engineering in Medicine and Biology, 2010
In this paper we present a novel hybrid approach for multimodal medical image registration based on diffeomorphic demons. Diffeomorphic demons have proven to be a robust and efficient way for intensity-based image registration. A very recent extension even allows to use mutual information (MI) as a similarity measure to registration multimodal images. However, due to the intensity correspondence uncertainty existing in some anatomical parts, it is difficult for a purely intensitybased algorithm to solve the registration problem. Therefore, we propose to combine the resulting transformations from both intensity-based and landmark-based methods for multimodal non-rigid registration based on diffeomorphic demons. Several experiments on different types of MR images were conducted, for which we show that a better anatomical correspondence between the images can be obtained using the hybrid approach than using either intensity information or landmarks alone.
Deformable Registration of Brain Tumor Images Via a Statistical Model of Tumor-Induced Deformation
2005
An approach to the deformable registration of three-dimensional brain tumor images to a normal brain atlas is presented. The approach involves the integration of three components: a biomechanical model of tumor mass-effect, a statistical approach to estimate the model's parameters, and a deformable image registration method. Statistical properties of the sought deformation map from the atlas to the image of a tumor patient are first obtained through tumor mass-effect simulations on normal brain images. This map is decomposed into the sum of two components in orthogonal subspaces, one representing inter-individual differences in brain shape, and the other representing tumor-induced deformation. For a new tumor case, a partial observation of the sought deformation map is obtained via deformable image registration and is decomposed into the aforementioned spaces in order to estimate the mass-effect model parameters. Using this estimate, a simulation of tumor mass-effect is performed on the atlas image in order to generate an image that is similar to tumor patient's image, thereby facilitating the atlas registration process. Results for a real tumor case and a number of simulated tumor cases indicate significant reduction in the registration error due to the presented approach as compared to the direct use of deformable image registration.
3rd IEEE International Symposium on Biomedical Imaging: Macro to Nano, 2006., 2006
A deformable registration method is proposed to register a brain atlas with tumor-bearing brain scans. The tumor mass effect is first simulated in the (normal) atlas, using a biomechanical model of mass effect. The tumor-bearing atlas is subsequently warped to the patient's scan by a deformable registration method, built upon the idea of HAMMER registration algorithm developed for normal brains. The potential of using the pattern of deformation around the tumor region to optimize the location of tumor seed and other parameters of the tumor model is also explored. Quantitative evaluation on simulated data shows that the proposed method achieves accuracy similar to that achieved in registration of images without tumors. Moreover, limited registration results on real tumors are promising.
Non-rigid Multimodal Image Registration based on the Expectation-Maximization Algorithm
In this paper, we present a novel methodology for multi-modal non-rigid image registration. The proposed approach is formulated by using the Expectation-Maximization (EM) technique in order to estimate a displacement vector field that aligns the images to register. In this approach, the image alignment relies on hidden stochastic random variables which allow to compare the intensity values between images of different modality. The methodology is basically composed of two steps: first, we provide an initial estimation of the the global deformation vector field by using a rigid registration technique based on particle filtering, obtaining, at the same time, an initial estimation of the joint conditional intensity distribution of the registered images; second, we approximate the remaining deformations by applying an iterative EM-technique approach , where at each step, a new estimation of the joint conditional intensity distribution and the displacement vector field are computed. The proposed algorithm was tested with different kinds of medical images ; preliminary results show that the methodology is a good alternative for non-rigid multimodal registration.