Three-dimensional anatomical model-based segmentation of MR brain images through Principal Axes Registration

A Fast Subpixel Registration Algorithm Based on Single-Step DFT Combined with Phase Correlation Constraint in Multimodality Brain Image

Multimodality brain image registration technology is the key technology to determine the accuracy and speed of brain diagnosis and treatment. In order to achieve high-precision image registration, a fast subpixel registration algorithm based on single-step DFT combined with phase correlation constraint in multimodality brain image was proposed in this paper. Firstly, the coarse positioning at the pixel level was achieved by using the downsampling cross-correlation model, which reduced the Fourier transform dimension of the cross-correlation matrix and the multiplication of the discrete Fourier transform matrix, so as to speed up the coarse registration process. Then, the improved DFT multiplier of the matrix multiplication was used in the neighborhood of the coarse point, and the subpixel fast location was achieved by the bidirectional search strategy. Qualitative and quantitative simulation experiment results show that, compared with comparison registration algorithms, our proposed algorithm could greatly reduce space and time complexity without losing accuracy.

Multitemporal Volume Registration for the Analysis of Rheumatoid Arthritis Evolution in the Wrist

This paper describes a method based on an automatic segmentation process to coregister carpal bones of the same patient imaged at different time points. A rigid registration was chosen to avoid artificial bone deformations and to allow finding eventual differences in the bone shape due to erosion, disease regression, or other eventual pathological signs. The actual registration step is performed on the basis of principal inertial axes of each carpal bone volume, as estimated from the inertia matrix. In contrast to already published approaches, the proposed method suggests splitting the 3D rotation into successive rotations about one axis at a time (the so-called basic or elemental rotations). In such a way, singularity and ambiguity drawbacks affecting other classical methods, for instance, the Euler angles method, are addressed. The proposed method was quantitatively evaluated using a set of real magnetic resonance imaging (MRI) sequences acquired at two different times from healthy wrists and by choosing a direct volumetric comparison as a cost function. Both the segmentation and registration steps are not based on a priori models, and they are therefore able to obtain good results even in pathological cases, as proven by the visual evaluation of actual pathological cases.

Fully-Automated μMRI Morphometric Phenotyping of the Tc1 Mouse Model of Down Syndrome

We describe a fully automated pipeline for the morphometric phenotyping of mouse brains from μMRI data, and show its application to the Tc1 mouse model of Down syndrome, to identify new morphological phenotypes in the brain of this first transchromosomic animal carrying human chromosome 21. We incorporate an accessible approach for simultaneously scanning multiple ex vivo brains, requiring only a 3D-printed brain holder, and novel image processing steps for their separation and orientation. We employ clinically established multi-atlas techniques–superior to single-atlas methods–together with publicly-available atlas databases for automatic skull-stripping and tissue segmentation, providing high-quality, subject-specific tissue maps. We follow these steps with group-wise registration, structural parcellation and both Voxel- and Tensor-Based Morphometry–advantageous for their ability to highlight morphological differences without the laborious delineation of regions of interest. We show the application of freely available open-source software developed for clinical MRI analysis to mouse brain data: NiftySeg for segmentation and NiftyReg for registration, and discuss atlases and parameters suitable for the preclinical paradigm. We used this pipeline to compare 29 Tc1 brains with 26 wild-type littermate controls, imaged ex vivo at 9.4T. We show an unexpected increase in Tc1 total intracranial volume and, controlling for this, local volume and grey matter density reductions in the Tc1 brain compared to the wild-types, most prominently in the cerebellum, in agreement with human DS and previous histological findings.

Computer-aided assessment of regional abdominal fat with food residue removal in CT

RATIONALE AND OBJECTIVES

Separate quantification of abdominal subcutaneous and visceral fat regions is essential to understand the role of regional adiposity as risk factor in epidemiological studies. Fat quantification is often based on computed tomography (CT) because fat density is distinct from other tissue densities in the abdomen. However, the presence of intestinal food residues with densities similar to fat may reduce fat quantification accuracy. We introduce an abdominal fat quantification method in CT with interest in food residue removal.

MATERIALS AND METHODS

Total fat was identified in the feature space of Hounsfield units and divided into subcutaneous and visceral components using model-based segmentation. Regions of food residues were identified and removed from visceral fat using a machine learning method integrating intensity, texture, and spatial information. Cost-weighting and bagging techniques were investigated to address class imbalance.

RESULTS

We validated our automated food residue removal technique against semimanual quantifications. Our feature selection experiments indicated that joint intensity and texture features produce the highest classification accuracy at 95%. We explored generalization capability using k-fold cross-validation and receiver operating characteristic (ROC) analysis with variable k. Losses in accuracy and area under ROC curve between maximum and minimum k were limited to 0.1% and 0.3%. We validated tissue segmentation against reference semimanual delineations. The Dice similarity scores were as high as 93.1 for subcutaneous fat and 85.6 for visceral fat.

CONCLUSIONS

Computer-aided regional abdominal fat quantification is a reliable computational tool for large-scale epidemiological studies. Our proposed intestinal food residue reduction scheme is an original contribution of this work. Validation experiments indicate very good accuracy and generalization capability.

Automatic recognition of surface landmarks of anatomical structures of back and posture

Faulty postures, scoliosis and sagittal plane deformities should be detected as early as possible to apply preventive and treatment measures against major clinical consequences. To support documentation of the severity of deformity and diminish x-ray exposures, several solutions utilizing analysis of back surface topography data were introduced. A novel approach to automatic recognition and localization of anatomical landmarks of the human back is presented that may provide more repeatable results and speed up the whole procedure. The algorithm was designed as a two-step process involving a statistical model built upon expert knowledge and analysis of three-dimensional back surface shape data. Voronoi diagram is used to connect mean geometric relations, which provide a first approximation of the positions, with surface curvature distribution, which further guides the recognition process and gives final locations of landmarks. Positions obtained using the developed algorithms are validated with respect to accuracy of manual landmark indication by experts. Preliminary validation proved that the landmarks were localized correctly, with accuracy depending mostly on the characteristics of a given structure. It was concluded that recognition should mainly take into account the shape of the back surface, putting as little emphasis on the statistical approximation as possible.

RIGID MEDICAL IMAGE REGISTRATION AND ITS ASSOCIATION WITH MUTUAL INFORMATION

Image registration plays a crucial role in the computer vision and medical imaging field where it is used to develop a spatial mapping between different sets of data. These transformations can range from simple rigid registrations to complex nonrigid deformations. Mutual information (MI) is a popular entropy-based similarity measure which has recently experienced a prolific expansion in a number of image registration applications. Stemming from information theory, this measure generally outperforms most other intensity-based measures in multimodal applications as it only assumes a statistical dependence between images. This paper provides a thorough introduction to the MI measure and its use in rigid medical image registration. A look at the extensions proposed to the original measure will also be provided. These were developed to improve the robustness of the measure and to avoid certain cases when maximizing MI does not lead to the correct spatial alignment.

Towards automatic quantification of the epicardial fat in non-contrasted CT images

In this work, we present a technique to semi-automatically quantify the epicardial fat in non-contrasted computed tomography (CT) images. The epicardial fat is very close to the pericardial fat, being separated only by the pericardium that appears in the image as a very thin line, which is hard to detect. Therefore, an algorithm that uses the anatomy of the heart was developed to detect the pericardium line via control points of the line. From the points detected an interpolation was applied based on the cubic interpolation, which was also improved to avoid incorrect interpolation that occurs when the two variables are non-monotonic. The method is validated by using a set of 40 CT images of the heart of 40 human subjects. In 62.5% of the cases only minimal user intervention was required and the results compared favourably with the results obtained by the manual process.

Image registration using hierarchical B-splines

Hierarchical B-splines have been widely used for shape modeling since their discovery by Forsey and Bartels. In this paper, we present an application of this concept, in the form of free-form deformation, to image registration by matching two images at increasing levels of detail. Results using MRI brain data are presented that demonstrate high degrees of matching while unnecessary distortions are avoided. We compare our results with the nonlinear ICP (Iterative Closest Point) algorithm (used for landmark-based registration) and optical flow (used for intensity-based registration).

Design and evaluation of an automatic procedure for detection of large misregistration of medical images

In many cases the combined assessment of three-dimensional anatomical and functional images [single photon emission computed tomography (SPECT), positron emission tomography (PET), magnetic resonance imaging (MRI), and computed tomography (CT)] is necessary to determine the precise nature and extent of lesions. It is important, prior to performing the addition, subtraction, or any other combination of the images, that they be adequately aligned and registered either by experienced radiologists via visual inspection, mental reorientation and overlap of slices, or by an automated registration algorithm. To be useful clinically, the latter case requires validation. The human capacity to evaluate registration results visually is limited and time consuming. This paper describes an algorithmic procedure to provide proxy measures for human assessment that discriminate between badly misregistered pairs of brain images and those likely to be clinically useful. The new algorithm consists of four major steps: segmentation of brain and skin/air boundaries, contour extraction, computation of the principal axes, and computation of the registration quality measures from the contour volumes. The test data were MR and CT brain images. The results of the present study indicate that the use of a measure based on the combination of brain and skin contours and a principal axis function is a good first step to reduce the number of badly registered images reaching the clinician.

Image fusion in open-architecture PACS-environment

Multimodal digital imaging is common in many fields of diagnosis and therapy planning - there is great interest in matching globally, fusing or registering data from the same part of the body. In practice, there are still difficulties in customizing image fusion in hospitals. Efficient routine use of image fusion requires, among others, an image management infrastructure - a picture archiving and communication system (PACS) - to provide storage of image data in a standard digital format, intelligent image management and fault-tolerant high-speed image networking. In order to customize image fusion, advances in both fusion software and hardware are also needed. The algorithms should be automatic, fast and accurate enough. Registration of multimodal data also creates a need for different display techniques and user-friendly interfaces. Image fusion has been impractical and too tedious to be performed in routine work, but in the future, fused images will be used in clinical practice - even in teleradiological consultation.

Alignment of magnetic-resonance brain datasets with the stereotactical coordinate system

Neuroanatomical and neurofunctional studies are often referenced to high-resolution magnetic-resonance brain datasets. For the analysis of the cortical surface, mapping of functional information on to the cortex or visualization, it is necessary to remove the outer surfaces of the brain. For intersubject comparison, it is useful to align the dataset with a coordinate system and introduce a spatial normalization. We describe an image processing chain that combines all of these steps in an interaction-free procedure. We report on a period of 2 years of routine application of this procedure, with >250 successfully processed datasets from healthy subjects and patients with various forms of brain damage.

Knowledge-based method for segmentation and analysis of lung boundaries in chest X-ray images

We present a knowledge-based approach to segmentation and analysis of the lung boundaries in chest X-rays. Image edges are matched to an anatomical model of the lung boundary using parametric features. A modular system architecture was developed which incorporates the model, image processing routines, an inference engine and a blackboard. Edges associated with the lung boundary are automatically identified and abnormal features are reported. In preliminary testing on 14 images for a set of 18 detectable abnormalities, the system showed a sensitivity of 88% and a specificity of 95% when compared with assessment by an experienced radiologist.

Registration of MR/MR and MR/SPECT brain images by fast stochastic optimization of robust voxel similarity measures

This paper describes a robust, fully automated algorithm to register intrasubject 3D single and multimodal images of the human brain. The proposed technique accounts for the major limitations of the existing voxel similarity-based methods: sensitivity of the registration to local minima of the similarity function and inability to cope with gross dissimilarities in the two images to be registered. Local minima are avoided by the implementation of a stochastic iterative optimization technique (fast simulated annealing). In addition, robust estimation is applied to reject outliers in case the images show significant differences (due to lesion evolution, incomplete acquisition, non-Gaussian noise, etc.). In order to evaluate the performance of this technique, 2D and 3D MR and SPECT human brain images were artificially rotated, translated, and corrupted by noise. A test object was acquired under different angles and positions for evaluating the accuracy of the registration. The approach has also been validated on real multiple sclerosis MR images of the same patient taken at different times. Furthermore, robust MR/SPECT image registration has permitted the representation of functional features for patients with partially complex seizures. The fast simulated annealing algorithm combined with robust estimation yields registration errors that are less than 1 degree in rotation and less than 1 voxel in translation (image dimensions of 128(3)). It compares favorably with other standard voxel similarity-based approaches.

Iterative Principal Axes Registration method for analysis of MR-PET brain images

Computerized automatic registration of MR-PET images of the brain is of significant interest for multimodality brain image analysis. In this paper, we discuss the Principal Axes Transformation for registration of three-dimensional MR and PET images. A new brain phantom designed to test MR-PET registration accuracy determines that the Principal Axes Registration method is accurate to within an average of 1.37 mm with a standard deviation of 0.78 mm. Often the PET scans are not complete in the sense that the PET volume does not match the respective MR volume. We have developed an Iterative Principal Axes Registration (IPAR) algorithm for such cases. Partial volumes of PET can be accurately registered to the complete MR volume using the new iterative algorithm. The quantitative and qualitative analyses of MR-PET image registration are presented and discussed. Results show that the new Principal Axes Registration algorithm is accurate and practical in MR-PET correlation studies.