scholarly journals Incorporating Metric Flows and Sparse Jacobian Transformations in ITK

2006 ◽  
Author(s):  
Mathieu De craene ◽  
Aloys Du bois d'aische ◽  
Benoit Macq ◽  
Simon Warfield
Keyword(s):  
Image Registration ◽  
Mutual Information ◽  
Computation Time ◽  
Deformable Registration ◽  
Deformation Field ◽  
Transformation Model ◽  
Rigid Registration ◽  
Information Metric ◽  
Demons Algorithm

Various metrics have been proposed in the literature for performing intrinsic automatic image to image registration. Among these measures, mutual information is a very popular one because of its robustness and accuracy for a wide variety of applications. In this paper, we propose a filter for performing non-rigid registration by estimating a dense deformation field derived from the mutual information metric. This filter takes place in the ITK PDE deformable registration design like the Demons algorithm of Thirion. We also show how the concept of metric flow is conceptually linked to the concept of metric derivative for a prior transformation model by the transformation jacobian. We also suggest a sparse implementation of the GetJacobian() method for reducing the computation time of a metric derivative for local transformations models.

Computation of Mutual Information Metric for Image Registration on Multiple GPUs

2014 ◽  
pp. 208-217
Author(s):  
Andrew Adinetz ◽  
Jiri Kraus ◽  
Markus Axer ◽  
Marcel Huysegoms ◽  
Stefan Köhnen ◽  
...  
Keyword(s):  
Image Registration ◽  
Mutual Information ◽  
Multiple Gpus ◽  
Information Metric

GRID-ENABLED NON-RIGID REGISTRATION OF MEDICAL IMAGES

2004 ◽  
Vol 14 (02) ◽  
pp. 197-216 ◽  
Author(s):  
RADU STEFANESCU ◽  
XAVIER PENNEC ◽  
NICHOLAS AYACHE
Keyword(s):  
High Performance ◽  
Cluster Computing ◽  
Parallel Implementation ◽  
Low Cost ◽  
Computation Time ◽  
Clinical Environment ◽  
Healthcare Organizations ◽  
Rigid Registration ◽  
Demons Algorithm ◽  
Point To Point

Over recent years, non-rigid registration has become a major issue in medical imaging. It consists in recovering a dense point-to-point correspondence field between two images, and usually takes a long time. This is in contrast to the needs of a clinical environment, where usability and speed are major constraints, leading to the necessity of reducing the computation time from slightly less than an hour to just a few minutes. As financial pressure makes it hard for healthcare organizations to invest in expensive high-performance computing (HPC) solutions, cluster computing proves to be a convenient solution to our computation needs, offering a large processing power at a low cost. Among the fast and efficient non-rigid registration methods, we chose the demons algorithm for its simplicity and good performances. The parallel implementation decomposes the correspondence field into spatial blocks, each block being assigned to a node of the cluster. We obtained an acceleration of 11 by using 15 2GHz PC's connected through a 1GB/s Ethernet network and reduced the computation time from 40min to 3min30. In order to further optimize the costs and the maintenance load, we investigate in the second part the transparent use of shared computing resources, either through a graphic client or a Web one.

scholarly journals Mjolnir: Extending HAMMER Using a Diffusion Transformation Model and Histogram Equalization for Deformable Image Registration

2009 ◽  
Vol 2009 ◽  
pp. 1-18 ◽  
Author(s):  
Lotta M. Ellingsen ◽  
Jerry L. Prince
Keyword(s):  
Image Registration ◽  
Medical Image Analysis ◽  
Computation Time ◽  
Deformable Image Registration ◽  
Histogram Equalization ◽  
Image Features ◽  
Transformation Model ◽  
Brain Images ◽  
Novel Approach ◽  
Feature Correspondences

Image registration is a crucial step in many medical image analysis procedures such as image fusion, surgical planning, segmentation and labeling, and shape comparison in population or longitudinal studies. A new approach to volumetric intersubject deformable image registration is presented. The method, called Mjolnir, is an extension of the highly successful method HAMMER. New image features in order to better localize points of correspondence between the two images are introduced as well as a novel approach to generate a dense displacement field based upon the weighted diffusion of automatically derived feature correspondences. An extensive validation of the algorithm was performed on T1-weighted SPGR MR brain images from the NIREP evaluation database. The results were compared with results generated by HAMMER and are shown to yield significant improvements in cortical alignment as well as reduced computation time.

An Improved Demon Registration with Mutual Information for Non-Rigid Medical Images

2013 ◽  
Vol 284-287 ◽  
pp. 1622-1626 ◽  
Author(s):  
Xiao Hui Xie ◽  
Cui Ma ◽  
Qiang Sun ◽  
Ru Xu Du
Keyword(s):  
Image Analysis ◽  
Image Registration ◽  
Mutual Information ◽  
Critical Points ◽  
Medical Image ◽  
Medical Image Analysis ◽  
Good Method ◽  
Demons Algorithm ◽  
Single Modality ◽  
Ct And Mri

Non-rigid image registration plays an important role in medical imaging. Classic Demons algorithm is a good method for image registration in some domain. One disadvantage of classic Demons algorithm is that the topological preservation can not be ensured, and it can only adapt to deal with the single modality image registration. In medical image analysis, the different modal images comparison and fusion are needed to give the doctor enough information for making a decision. The mutual information algorithm has been validated useful for multi-modality image registration. By analyzing the critical points of Demons registration like mis-registration, an improved Demons algorithm with mutual information evaluation is proposed. Experiment results on liver images between CT and MRI modality show that the proposed algorithm can deal with multi-modality image registration well and it can hold the abilities even faces the noise and distortion.

scholarly journals Complex Dual Tree Wavelet Multi-Level Feature Based Transformation Parameter Estimation for 3d Medical Image Registration

2019 ◽  
Vol 8 (9) ◽  
pp. 799-806
Keyword(s):  
Image Registration ◽  
Mutual Information ◽  
3D Images ◽  
Low Pass ◽  
Feature Based ◽  
Information Metric ◽  
Transformation Parameters ◽  
3D Image Registration ◽  
High Pass ◽  
Ct And Mri

3D image registration of CT and MRI data is carried out using DTCWT sub bands by considering the features from all 64 bands. The features are selected by considering Mattes Mutual Information Metric and the optimizer algorithm estimates the optimum transformation parameters from all the 64 bands. Transformation parameters from eight low pass bands from each octave are averaged to identify optimum registration parameters. Similarly, for registration of high pass bands mean of transformation parameters from 56 bands are identified. The proposed registration algorithm is suitable for register multimodal medical images and the proposed algorithm is validated for more than 20 3D images. Mutual information and joint entropy is estimated to demonstrate the advantages of proposed algorithm overt that of intensity based algorithm. With features identified from 56 bands with six orientations the registered image is found to consist of features from both input images with closeness level measured to be within 12%.

MULTI-MODALITY IMAGE REGISTRATION FOR SUBDURAL ELECTRODE LOCALIZATION

2014 ◽  
Vol 26 (05) ◽  
pp. 1450051
Author(s):  
Shuo Dong ◽  
Yuan Liu ◽  
Lixin Cai ◽  
Mei Bai ◽  
Hanmin Yan
Keyword(s):  
Image Registration ◽  
Intractable Epilepsy ◽  
Rigid Registration ◽  
Multimodal Image Registration ◽  
Subdural Electrodes ◽  
Subdural Electrode ◽  
Information Metric ◽  
Electrode Localization ◽  
Image Pairs

Surgical treatment has been proved to be an effective way to control seizures for some kinds of intractable epilepsy. The electrocorticogram (ECoG) recorded from subdural electrodes has become an important technique for defining epileptogenic zones before surgery in clinical practice. The exact location of subdural electrodes has to be determined to establish the connection between electrodes and epileptogenic zones. Artifacts caused by the electrodes can severely affect the quality of CT imaging and sequentially image registration. In this paper, we discussed the performance of mean squares and the Mattes mutual information metric methods in multimodal image registration for subdural electrode localization. Since the skull can be regarded as a rigid body, rigid registration is sufficient for the purpose of subdural electrode localization. The vital parameter for the rigid registration is rotation. The translation result depends on the result of rotation. Both the methods performed well in the determination of the rotation center. Rotation angles of different image pairs of the same volume pair fluctuated a lot. Based on the image acquisition process, we assume that the images within the same volume pair should have the same transformation parameters for registration. Results show that the mean rotation angles of images within one dataset are approximate to the manual results that are considered to be the actual result for registration despite their fluctuation range.

Rigid Medical Image Registration Based on Genetic Algorithms and Mutual Information

2014 ◽  
Vol 665 ◽  
pp. 712-717
Author(s):  
Wei Wei ◽  
Liang Liu ◽  
Zhong Qin Hu ◽  
Yu Jing Zhou
Keyword(s):  
Genetic Algorithm ◽  
Genetic Algorithms ◽  
Image Registration ◽  
Mutual Information ◽  
Medical Image ◽  
Medical Image Registration ◽  
Rigid Registration ◽  
The Matrix ◽  
Depth Study ◽  
Stability And Accuracy

With the variety of medical imaging equipment’s application in the medical process,medical image registration becomes particularly important in the field of medical image processing,which has important clinical diagnostic and therapeutic value. This article describes the matrix conversion method of the rigid registration model, the basic concepts and principles of the mutual information algorithm ,the basic idea of genetic algorithms and algorithm’s flow , and the application of the improved genetic algorithms in practice. The rigid registration of two CT brain bones images uses mutual information as a similarity measure, genetic algorithm as the search strategy and matlab as programming environment. Using the three-point crossover technique to exchange the three parameters in the rigid transformation repeectively to produce new individuals, the genetic algorithm’s local search ability enhanced and the prematurity phenomenon can be reduced through the depth study of the basic genetic algorithm. The experiments show that the registration has high stability and accuracy.

Sign in / Sign up

Export Citation Format

Share Document