Adaptive segmentation of MRI data

Automatic segmentation of the human brain ventricles from MR images by knowledge-based region growing and trimming

Automatic segmentation of the human brain ventricular system from MR images is useful in studies of brain anatomy and its diseases. Existing intensity-based segmentation methods are adaptive to large shape and size variations of the ventricular system, but may leak to the non-ventricular regions due to the non-homogeneity, noise and partial volume effect in the images. Deformable model-based methods are more robust to noise and alleviate the leakage problem, but may generate wrong results when the shape or size of the ventricle to be segmented in the images has a large difference in comparison to its model. In this paper, we propose a knowledge-based region growing and trimming approach where: (1) a model of a ventricular system is used to define regions of interest (ROI) for the four ventricles (i.e., left, right, third and fourth); (2) to segment a ventricle in its ROI, a region growing procedure is first applied to obtain a connected region that contains the ventricle, and (3) a region trimming procedure is then employed to trim the non-ventricle regions. A hysteretic thresholding is developed for the region growing procedure to cope with the partial volume effect and minimize non-ventricular regions. The domain knowledge on the shape and intensity features of the ventricular system is used for the region trimming procedure. Due to the joint use of the model-based and intensity-based approaches, our method is robust to noise and large shape and size variations. Experiments on 18 simulated and 58 clinical MR images show that the proposed approach is able to segment the ventricular system accurately with the dice similarity coefficient ranging from 91% to 99%.

Automated brain tumor segmentation using spatial accuracy-weighted hidden Markov Random Field

A variety of algorithms have been proposed for brain tumor segmentation from multi-channel sequences, however, most of them require isotropic or pseudo-isotropic resolution of the MR images. Although co-registration and interpolation of low-resolution sequences, such as T2-weighted images, onto the space of the high-resolution image, such as T1-weighted image, can be performed prior to the segmentation, the results are usually limited by partial volume effects due to interpolation of low-resolution images. To improve the quality of tumor segmentation in clinical applications where low-resolution sequences are commonly used together with high-resolution images, we propose the algorithm based on Spatial accuracy-weighted Hidden Markov random field and Expectation maximization (SHE) approach for both automated tumor and enhanced-tumor segmentation. SHE incorporates the spatial interpolation accuracy of low-resolution images into the optimization procedure of the Hidden Markov Random Field (HMRF) to segment tumor using multi-channel MR images with different resolutions, e.g., high-resolution T1-weighted and low-resolution T2-weighted images. In experiments, we evaluated this algorithm using a set of simulated multi-channel brain MR images with known ground-truth tissue segmentation and also applied it to a dataset of MR images obtained during clinical trials of brain tumor chemotherapy. The results show that more accurate tumor segmentation results can be obtained by comparing with conventional multi-channel segmentation algorithms.

Intensity correction with a pair of spoiled gradient recalled echo images

Intensity inhomogeneities in magnetic resonance images (MRI) are a frequently occurring artefact, and result in the same tissue class to have vastly different intensities within an image. These inhomogeneities can be modelled by a slowly varying field, which is also called the bias field. Previous phantom-, image- or sequence based approaches suffer from long scan times, post-processing times or do not sufficiently remove the intensity variations. These intensity variations cause problems for quantitative image analysis algorithms (segmentation, registration) as well as clinicians (e.g. by complicating the visual assessment). This paper presents a novel technique (COIN, correction of intensity inhomogeneities) that uses two calibration images (fast spoiled gradient echo) to map a parameter containing the bias field, which is specific to the patient during a particular exam. This parametric map can then be used to correct any other images acquired during the same exam, regardless of the sequence employed. By using a short repetition time (less than 5 ms) for the calibration scans, the additional scan time is reduced to 60 s (max). The subsequent post-processing time is approximately 60 s per 20 slices. We successfully validate our approach on simulated brain MRI as well as real liver and spinal images. These images were acquired with a number of different coils, sequences and weightings. A comparison of our method with an existing, commercially available algorithm by radiologists shows that COIN is superior.

Shape abnormalities of caudate nucleus in schizotypal personality disorder

BACKGROUND

Previously, we reported abnormal volume and global shape in the caudate nucleus in schizotypal personality disorder (SPD). Here, we use a new shape measure which importantly permits local in addition to global shape analysis, as well as local correlations with behavioral measures.

METHODS

Thirty-two female and 15 male SPDs, and 29 female and 14 male normal controls (NCLs), underwent brain magnetic resonance imaging (MRI). We assessed caudate shape measures using spherical harmonic-point distribution model (SPHARM-PDM) methodology.

RESULTS

We found more pronounced global shape differences in the right caudate in male and female SPD, compared with NCLs. Local shape differences, principally in the caudate head, survived statistical correction on the right. Also, we performed correlations between local surface deformations with clinical measures and found significant correlations between local shape deflated deformations in the anterior medial surface of the caudate with verbal learning capacity in female SPD.

CONCLUSIONS

Using SPHARM-PDM methodology, we found both global and local caudate shape abnormalities in male and female SPD, particularly right-sided, and largely restricted to limbic and cognitive anterior caudate. The most important and novel findings were bilateral statistically significant correlations between local surface deflations in the anterior medial surface of the head of the caudate and verbal learning capacity in female SPD. By extension, these local caudate correlation findings implicate the ventromedial prefrontal cortex (vmPFC), which innervates that area of the caudate, and demonstrate the utility of local shape analysis to investigate the relationship between specific subcortical and cortical brain structures in neuropsychiatric conditions.

A hybrid geometric-statistical deformable model for automated 3-D segmentation in brain MRI

We present a novel 3-D deformable model-based approach for accurate, robust, and automated tissue segmentation of brain MRI data of single as well as multiple magnetic resonance sequences. The main contribution of this study is that we employ an edge-based geodesic active contour for the segmentation task by integrating both image edge geometry and voxel statistical homogeneity into a novel hybrid geometric-statistical feature to regularize contour convergence and extract complex anatomical structures. We validate the accuracy of the segmentation results on simulated brain MRI scans of both single T1-weighted and multiple T1/T2/PD-weighted sequences. We also demonstrate the robustness of the proposed method when applied to clinical brain MRI scans. When compared to a current state-of-the-art region-based level-set segmentation formulation, our white matter and gray matter segmentation resulted in significantly higher accuracy levels with a mean improvement in Dice similarity indexes of 8.55% ( p < 0.0001) and 10.18% ( p < 0.0001), respectively.

The emerging discipline of Computational Functional Anatomy

Computational Functional Anatomy (CFA) is the study of functional and physiological response variables in anatomical coordinates. For this we focus on two things: (i) the construction of bijections (via diffeomorphisms) between the coordinatized manifolds of human anatomy, and (ii) the transfer (group action and parallel transport) of functional information into anatomical atlases via these bijections. We review advances in the unification of the bijective comparison of anatomical submanifolds via point-sets including points, curves and surface triangulations as well as dense imagery. We examine the transfer via these bijections of functional response variables into anatomical coordinates via group action on scalars and matrices in DTI as well as parallel transport of metric information across multiple templates which preserves the inner product.

Inhomogeneity compensation for MR brain image segmentation using a multi-stage FCM-based approach

Intensity inhomogeneity or intensity non-uniformity (INU) is an undesired phenomenon that represents the main obstacle for MR image segmentation and registration methods. Various techniques have been proposed to eliminate or compensate the INU, most of which are embedded into clustering algorithms. This paper proposes a multiple stage fuzzy c-means (FCM) based algorithm for the estimation and compensation of the slowly varying additive or multiplicative noise, supported by a pre-filtering technique for Gaussian and impulse noise elimination. The slowly varying behavior of the bias or gain field is assured by a smoothening filter that performs a context dependent averaging, based on a morphological criterion. The experiments using 2-D synthetic phantoms and real MR images show, that the proposed method provides accurate segmentation. The produced segmentation and fuzzy membership values can serve as excellent support for 3-D registration and segmentation techniques.

Segmentation of diffusion-weighted brain images using expectation maximization algorithm initialized by hierarchical clustering

Tissue segmentation based on diffusion-weighted images (DWI) provides complementary information of tissue contrast to the structural MRI for facilitating the tissue segmentation. In the previous literatures, DWI-based brain tissue segmentation was carried out using the parametric images, such as fractional anisotropy (FA) and apparent diffusion coefficient (ADC). However, the information of directions of neural fibers was very limited in the parametric images. To fully utilize the directional information, we propose a novel method to perform tissue segmentation directly on the DWI raw image data. Specifically, a hierarchical clustering (HC) technique was first applied on the down-sampled data to initialize the model parameters for each tissue cluster followed by automatic segmentation using the expectation maximization (EM) algorithm. The whole brain DWI raw data of five normal subjects were analyzed. The results demonstrated that HC-EM is effective in multi-tissue classification on DWI raw data.

Nonrigid registration of joint histograms for intensity standardization in magnetic resonance imaging

A major disadvantage of magnetic resonance imaging (MRI) compared to other imaging modalities like computed tomography is the fact that its intensities are not standardized. Our contribution is a novel method for MRI signal intensity standardization of arbitrary MRI scans, so as to create a pulse sequence dependent standard intensity scale. The proposed method is the first approach that uses the properties of all acquired images jointly (e.g., T1- and T2-weighted images). The image properties are stored in multidimensional joint histograms. In order to normalize the probability density function (pdf) of a newly acquired data set, a nonrigid image registration is performed between a reference and the joint histogram of the acquired images. From this matching a nonparametric transformation is obtained, which describes a mapping between the corresponding intensity spaces and subsequently adapts the image properties of the newly acquired series to a given standard. As the proposed intensity standardization is based on the probability density functions of the data sets only, it is independent of spatial coherence or prior segmentations of the reference and current images. Furthermore, it is not designed for a particular application, body region or acquisition protocol. The evaluation was done using two different settings. First, MRI head images were used, hence the approach can be compared to state-of-the-art methods. Second, whole body MRI scans were used. For this modality no other normalization algorithm is known in literature. The Jeffrey divergence of the pdfs of the whole body scans was reduced by 45%. All used data sets were acquired during clinical routine and thus included pathologies.

MRI tissue classification and bias field estimation based on coherent local intensity clustering: a unified energy minimization framework

This paper presents a new energy minimization method for simultaneous tissue classification and bias field estimation of magnetic resonance (MR) images. We first derive an important characteristic of local image intensities--the intensities of different tissues within a neighborhood form separable clusters, and the center of each cluster can be well approximated by the product of the bias within the neighborhood and a tissue-dependent constant. We then introduce a coherent local intensity clustering (CLIC) criterion function as a metric to evaluate tissue classification and bias field estimation. An integration of this metric defines an energy on a bias field, membership functions of the tissues, and the parameters that approximate the true signal from the corresponding tissues. Thus, tissue classification and bias field estimation are simultaneously achieved by minimizing this energy. The smoothness of the derived optimal bias field is ensured by the spatially coherent nature of the CLIC criterion function. As a result, no extra effort is needed to smooth the bias field in our method. Moreover, the proposed algorithm is robust to the choice of initial conditions, thereby allowing fully automatic applications. Our algorithm has been applied to high field and ultra high field MR images with promising results.

Model-based segmentation of hippocampal subfields in ultra-high resolution in vivo MRI

Recent developments in MR data acquisition technology are starting to yield images that show anatomical features of the hippocampal formation at an unprecedented level of detail, providing the basis for hippocampal subfield measurement. Because of the role of the hippocampus in human memory and its implication in a variety of disorders and conditions, the ability to reliably and efficiently quantify its subfields through in vivo neuroimaging is of great interest to both basic neuroscience and clinical research. In this paper, we propose a fully-automated method for segmenting the hippocampal subfields in ultra-high resolution MRI data. Using a Bayesian approach, we build a computational model of how images around the hippocampal area are generated, and use this model to obtain automated segmentations. We validate the proposed technique by comparing our segmentation results with corresponding manual delineations in ultra-high resolution MRI scans of five individuals.

Atlas-based segmentation of the germinal matrix from in utero clinical MRI of the fetal brain

Recently developed techniques for reconstruction of high-resolution 3D images from fetal MR scans allows us to study the morphometry of developing brain tissues in utero. However, existing adult brain analysis methods cannot be directly applied as the anatomy of the fetal brain is significantly different in terms of geometry and tissue morphology. We describe an approach to atlas-based segmentation of the fetal brain with particular focus on the delineation of the germinal matrix, a transient structure related to brain growth. We segment 3D images reconstructed from in utero clinical MR scans and measure volumes of different brain tissue classes for a group of fetal subjects at gestational age 20.5-22.5 weeks. We also include a partial validation of the approach using manual tracing of the germinal matrix at different gestational ages.

Image background inhomogeneity correction in MRI via intensity standardization

An automatic, simple, and image intensity standardization-based strategy for correcting background inhomogeneity in MR images is presented in this paper. Image intensities are first transformed to a standard intensity gray scale by a standardization process. Different tissue sample regions are then obtained from the standardized image by simply thresholding based on fixed intensity intervals. For each tissue region, a polynomial is fitted to the estimated discrete background intensity variation. Finally, a combined polynomial is determined and used for correcting the intensity inhomogeneity in the whole image. The above procedure is repeated on the corrected image iteratively until the size of the extracted tissue regions does not change significantly in two successive iterations. Intensity scale standardization is effected to make sure that the corrected image is not biased by the fitting strategy. The method has been tested on a number of simulated and clinical MR images. These tests and a comparison with the method of non-parametric non-uniform intensity normalization (N3) indicate that the method is effective in background intensity inhomogeneity correction and may have a slight edge over the N3 method.

Validation of alternating Kernel mixture method: application to tissue segmentation of cortical and subcortical structures

This paper describes the application of the alternating Kernel mixture (AKM) segmentation algorithm to high resolution MRI subvolumes acquired from a 1.5T scanner (hippocampus, n = 10 and prefrontal cortex, n = 9) and a 3T scanner (hippocampus, n = 10 and occipital lobe, n = 10). Segmentation of the subvolumes into cerebrospinal fluid, gray matter, and white matter tissue is validated by comparison with manual segmentation. When compared with other segmentation methods that use traditional Bayesian segmentation, AKM yields smaller errors (P < .005, exact Wilcoxon signed rank test) demonstrating the robustness and wide applicability of AKM across different structures. By generating multiple mixtures for each tissue compartment, AKM mimics the increased variation of manual segmentation in partial volumes due to the highly folded tissues. AKM's superior performance makes it useful for tissue segmentation of subcortical and cortical structures in large-scale neuroimaging studies.

Bias artifact suppression on MR volumes

RF-inhomogeneity correction is a relevant research topic in the field of magnetic resonance imaging (MRI). A volume corrupted by this artifact exhibits nonuniform illumination both inside a single slice and between adjacent ones. In this work a bias correction technique is presented, which suppresses this artifact on MR volumes scanned from different body parts without any a priori hypothesis on the artifact model. Theoretical foundations of the method are reported together with experimental results and a comparison is presented with both the 2D version of the algorithm and other techniques that are widely used in MRI literature.

An MRI study of age-related white and gray matter volume changes in the rhesus monkey

We applied the automated MRI segmentation technique Template Driven Segmentation (TDS) to dual-echo spin echo (DE SE) images of eight young (5-12 years), six middle-aged (16-19 years) and eight old (24-30 years) rhesus monkeys. We analyzed standardized mean volumes for 18 anatomically defined regions of interest (ROI's) and found an overall decrease from young to old age in the total forebrain (5.01%), forebrain parenchyma (5.24%), forebrain white matter (11.53%), forebrain gray matter (2.08%), caudate nucleus (11.79%) and globus pallidus (18.26%). Corresponding behavioral data for five of the young, five of the middle-aged and seven of the old subjects on the Delayed Non-matching to Sample (DNMS) task, the Delayed-recognition Span Task (DRST) and the Cognitive Impairment Index (CII) were also analyzed. We found that none of the cognitive measures were related to ROI volume changes in our sample size of monkeys.

Graph-partitioned spatial priors for functional magnetic resonance images

Spatial models of functional magnetic resonance imaging (fMRI) data allow one to estimate the spatial smoothness of general linear model (GLM) parameters and eschew pre-process smoothing of data entailed by conventional mass-univariate analyses. Recently diffusion-based spatial priors [Harrison, L.M., Penny, W., Daunizeau, J., and Friston, K.J. (2008). Diffusion-based spatial priors for functional magnetic resonance images. NeuroImage.] were proposed, which provide a way to formulate an adaptive spatial basis, where the diffusion kernel of a weighted graph-Laplacian (WGL) is used as the prior covariance matrix over GLM parameters. An advantage of these is that they can be used to relax the assumption of isotropy and stationarity implicit in smoothing data with a fixed Gaussian kernel. The limitation of diffusion-based models is purely computational, due to the large number of voxels in a brain volume. One solution is to partition a brain volume into slices, using a spatial model for each slice. This reduces computational burden by approximating the full WGL with a block diagonal form, where each block can be analysed separately. While fMRI data are collected in slices, the functional structures exhibiting spatial coherence and continuity are generally three-dimensional, calling for a more informed partition. We address this using the graph-Laplacian to divide a brain volume into sub-graphs, whose shape can be arbitrary. Their shape depends crucially on edge weights of the graph, which can be based on the Euclidean distance between voxels (isotropic) or on GLM parameters (anisotropic) encoding functional responses. The result is an approximation the full WGL that retains its 3D form and also has potential for parallelism. We applied the method to high-resolution (1 mm(3)) fMRI data and compared models where a volume was divided into either slices or graph-partitions. Models were optimized using Expectation-Maximization and the approximate log-evidence computed to compare these different ways to partition a spatial prior. The high-resolution fMRI data presented here had greatest evidence for the graph partitioned anisotropic model, which was best able to preserve fine functional detail.

Robust MRI brain tissue parameter estimation by multistage outlier rejection

This article addresses the problem of the tissue type parameter estimation in brain MRI in the presence of partial volume effects. Automatic MRI brain tissue classification is hampered by partial volume effects that are caused by the finite resolution of the acquisition process. Due to this effect intensity distributions in brain MRI cannot be well modeled by a simple mixture of Gaussians and therefore more complex models have been developed. Unfortunately, these models do not seem to be robust enough for clinical conditions, as the quality of the tissue classification decreases rapidly with the image quality. Also, the application of these methods for pathological images with unmodeled intensities (e.g. MS plaques, tumors, etc.) remains uncertain. In the present work a new robust method for brain tissue characterization is presented, treating the partial volume affected voxels as outliers of the pure tissue distributions. The proposed method estimates the tissue characteristics from a reduced set of intensities belonging to a particular pure tissue class. This reduced set is selected by using a trimming procedure based on local gradient information and distributional data. This feature makes the method highly tolerant of a large amount of unexpected intensities without degrading its performance. The proposed method has been evaluated using both synthetic and real MR data and compared with state-of-the-art methods showing the best results in the comparative.

Restoration of MRI data for intensity non-uniformities using local high order intensity statistics

MRI at high magnetic fields (>3.0 T) is complicated by strong inhomogeneous radio-frequency fields, sometimes termed the "bias field". These lead to non-biological intensity non-uniformities across the image. They can complicate further image analysis such as registration and tissue segmentation. Existing methods for intensity uniformity restoration have been optimized for 1.5 T, but they are less effective for 3.0 T MRI, and not at all satisfactory for higher fields. Also, many of the existing restoration algorithms require a brain template or use a prior atlas, which can restrict their practicalities. In this study an effective intensity uniformity restoration algorithm has been developed based on non-parametric statistics of high order local intensity co-occurrences. These statistics are restored with a non-stationary Wiener filter. The algorithm also assumes a smooth non-uniformity and is stable. It does not require a prior atlas and is robust to variations in anatomy. In geriatric brain imaging it is robust to variations such as enlarged ventricles and low contrast to noise ratio. The co-occurrence statistics improve robustness to whole head images with pronounced non-uniformities present in high field acquisitions. Its significantly improved performance and lower time requirements have been demonstrated by comparing it to the very commonly used N3 algorithm on BrainWeb MR simulator images as well as on real 4 T human head images.

A rhesus monkey reference label atlas for template driven segmentation

BACKGROUND

We have acquired dual-echo spin-echo (DE SE) MRI data of the rhesus monkey brain since 1994 as part of an ongoing study of normal aging. To analyze these legacy data for regional volume changes, we have created a reference label atlas for the Template Driven Segmentation (TDS) algorithm.

METHODS

The atlas was manually created from DE SE legacy MRI data of one behaviorally normal, young, male rhesus monkey and consisted of 14 regions of interest (ROI's). We analyzed the reproducibility and validity of the TDS algorithm using the atlas relative to manual segmentation.

RESULTS

ROI volumes were comparable between the two segmentation methodologies, except TDS overestimated the volume of basal ganglia regions. Both methodologies were highly reproducible, but TDS had lower sensitivity and comparable specificity.

CONCLUSIONS

TDS segmentation calculates accurate volumes for most ROI's. Sensitivity will be improved in future studies through the acquisition of higher quality data.

Feature-based fusion of medical imaging data

The acquisition of multiple brain imaging types for a given study is a very common practice. There have been a number of approaches proposed for combining or fusing multitask or multimodal information. These can be roughly divided into those that attempt to study convergence of multimodal imaging, for example, how function and structure are related in the same region of the brain, and those that attempt to study the complementary nature of modalities, for example, utilizing temporal EEG information and spatial functional magnetic resonance imaging information. Within each of these categories, one can attempt data integration (the use of one imaging modality to improve the results of another) or true data fusion (in which multiple modalities are utilized to inform one another). We review both approaches and present a recent computational approach that first preprocesses the data to compute features of interest. The features are then analyzed in a multivariate manner using independent component analysis. We describe the approach in detail and provide examples of how it has been used for different fusion tasks. We also propose a method for selecting which combination of modalities provides the greatest value in discriminating groups. Finally, we summarize and describe future research topics.

Fully automatic segmentation of multiple sclerosis lesions in brain MR FLAIR images using adaptive mixtures method and Markov random field model

In this paper, an approach is proposed for fully automatic segmentation of MS lesions in fluid attenuated inversion recovery (FLAIR) Magnetic Resonance (MR) images. The proposed approach, based on a Bayesian classifier, utilizes the adaptive mixtures method (AMM) and Markov random field (MRF) model to obtain and upgrade the class conditional probability density function (CCPDF) and the a priori probability of each class. To compare the performance of the proposed approach with those of previous approaches including manual segmentation, the similarity criteria of different slices related to 20 MS patients were calculated. Also, volumetric comparison of lesions volume between the fully automated segmentation and the gold standard was performed using correlation coefficient (CC). The results showed a better performance for the proposed approach, compared to those of previous works.

MR brain tissue classification using an edge-preserving spatially variant Bayesian mixture model

In this paper, a spatially constrained mixture model for the segmentation of MR brain images is presented. The novelty of this work is an edge-preserving smoothness prior which is imposed on the probabilities of the voxel labels. This prior incorporates a line process, which is modeled as a Bernoulli random variable, in order to preserve edges between tissues. The main difference with other, state of the art methods imposing priors, is that the constraint is imposed on the probabilities of the voxel labels and not onto the labels themselves. Inference of the proposed Bayesian model is obtained using variational methodology and the model parameters are computed in closed form. Numerical experiments are presented where the proposed model is favorably compared to state of the art brain segmentation methods as well as to a spatially varying Gaussian mixture model.

A vectorial image classification method based on neighborhood weighted Gaussian mixture model

The CT uroscan contains three to four time-spaced acquisitions of the same patient. Registration of these acquisitions forms a vectorial volume, which contains a more complete anatomical information. In order to outline the anatomical structures, multi-dimensional classification is necessary for analyzing this vectorial volume. Because of the partial volume effect (PVE), probability distributions are assigned to the different material types within this vectorial volume instead of a definite material distribution. Gaussian mixture model is often used in probability classification problems to model such distributions, but it relies only on the intensity distributions, which will lead a misclassification on the boundaries and inhomogeneous regions with noises. In order to solve this problem, a neighborhood weighted Gaussian mixture model is proposed in this paper. Expectation Maximization algorithm is used as optimization method. The experiments demonstrate that the proposed method can get a better classification result and less affected by the noise.

A variational level set approach to segmentation and bias correction of images with intensity inhomogeneity

This paper presents a variational level set approach to joint segmentation and bias correction of images with intensity inhomogeneity. Our method is based on an observation that intensities in a relatively small local region are separable, despite of the inseparability of the intensities in the whole image caused by the intensity inhomogeneity. We first define a weighted K-means clustering objective function for image intensities in a neighborhood around each point, with the cluster centers having a multiplicative factor that estimates the bias within the neighborhood. The objective function is then integrated over the entire domain and incorporated into a variational level set formulation. The energy minimization is performed via a level set evolution process. Our method is able to estimate bias of quite general profiles. Moreover, it is robust to initialization, and therefore allows automatic applications. The proposed method has been used for images of various modalities with promising results.

Assessment of reliability of multi-site neuroimaging via traveling phantom study

This paper describes a framework for quantitative analysis of neuroimaging data of traveling human phantoms used for cross-site validation. We focus on the analysis of magnetic resonance image data including intra- and inter-site comparison. Locations and magnitude of geometric deformation is studied via unbiased atlas building and metrics on deformation fields. Variability of tissue segmentation is analyzed by comparison of volumes, overlap of tissue maps, and a new Kullback-Leibler divergence on tissue probabilities, with emphasis on comparing probabilistic rather than binary segmentations. We show that results from this information theoretic measure are highly correlated with overlap. Reproducibility of automatic, atlas-based segmentation of subcortical structures is examined by comparison of volumes, shape overlap and surface distances. Variability among scanners of the same type but also differences to a different scanner type are discussed. The results demonstrate excellent reliability across multiple sites that can be achieved by the use of the today's scanner generation and powerful automatic analysis software. Knowledge about such variability is crucial for study design and power analysis in new multi-site clinical studies.

Entropy-controlled quadratic markov measure field models for efficient image segmentation

We present a new Markov random field (MRF) based model for parametric image segmentation. Instead of directly computing a label map, our method computes the probability that the observed data at each pixel is generated by a particular intensity model. Prior information about segmentation smoothness and low entropy of the probability distribution maps is codified in the form of a MRF with quadratic potentials so that the optimal estimator is obtained by solving a quadratic cost function with linear constraints. Although, for segmentation purposes, the mode of the probability distribution at each pixel is naturally used as an optimal estimator, our method permits the use of other estimators, such as the mean or the median, which may be more appropriate for certain applications. Numerical experiments and comparisons with other published schemes are performed, using both synthetic images and real data of brain MRI for which expert hand-made segmentations are available. Finally, we show that the proposed methodology may be easily extended to other problems, such as stereo disparity estimation.

Segmentation of brain parenchyma using bilateral filtering and region growing

When the non-diffusion weighted images (non-DWIs) and the diffusion weighted images (DWIs) are acquired by a fast imaging sequence, they suffer from several artifacts such as N/2 ghost, subject motion, eddy current, etc. These artifacts act as a noise in the background area of the human brain. To extract the brain region from the noisy background, brain parenchyma segmentation has been used. Several segmentation methods presented so far cannot address this problem well. In this study, we propose a novel segmentation method of brain contour in non-DWIs using bilateral filtering, which can reduce the background noise while edge-preserving, and region growing. We compare the segmentation results from various methods, and the proposed method shows better segmentation results than those from other schemes.

A fuzzy, nonparametric segmentation framework for DTI and MRI analysis: with applications to DTI-tract extraction

This paper presents a novel fuzzy-segmentation method for diffusion tensor (DT) and magnetic resonance (MR) images. Typical fuzzy-segmentation schemes, e.g., those based on fuzzy C means (FCM), incorporate Gaussian class models that are inherently biased towards ellipsoidal clusters characterized by a mean element and a covariance matrix. Tensors in fiber bundles, however, inherently lie on specific manifolds in Riemannian spaces. Unlike FCM-based schemes, the proposed method represents these manifolds using nonparametric data-driven statistical models. The paper describes a statistically-sound (consistent) technique for nonparametric modeling in Riemannian DT spaces. The proposed method produces an optimal fuzzy segmentation by maximizing a novel information-theoretic energy in a Markov-random-field framework. Results on synthetic and real, DT and MR images, show that the proposed method provides information about the uncertainties in the segmentation decisions, which stem from imaging artifacts including noise, partial voluming, and inhomogeneity. By enhancing the nonparametric model to capture the spatial continuity and structure of the fiber bundle, we exploit the framework to extract the cingulum fiber bundle. Typical tractography methods for tract delineation, incorporating thresholds on fractional anisotropy and fiber curvature to terminate tracking, can face serious problems arising from partial voluming and noise. For these reasons, tractography often fails to extract thin tracts with sharp changes in orientation, such as the cingulum. The results demonstrate that the proposed method extracts this structure significantly more accurately as compared to tractography.

Segmentation of brain MRI in young children

RATIONALE AND OBJECTIVES

This article deals with an automatic tissue segmentation of brain magnetic resonance imaging (MRI) in young children.

MATERIALS AND METHODS

We examine the suitability of state-of-the-art methods developed for the adult brain when applied to the segmentation of the brain MRI in young children. We develop a method of creation of a population-specific atlas in young children using a single manual segmentation. The method is based on nonlinear propagation of the segmentation into population and subsequent affine alignment into a reference space and averaging.

RESULTS

Using this approach, we significantly improve the performance of the popular expectation-maximization algorithm on brain MRI in young children. The method can be used for building probabilistic atlases with any number of structures. We compare resulting algorithm with nonrigid registration-based label propagation.

CONCLUSIONS

Finally, both methods are used to measure the volume of seven brain structures and measure the growth between 1 and 2 years of age.

Intensity non-uniformity correction using N3 on 3-T scanners with multichannel phased array coils

Measures of structural brain change based on longitudinal MR imaging are increasingly important but can be degraded by intensity non-uniformity. This non-uniformity can be more pronounced at higher field strengths, or when using multichannel receiver coils. We assessed the ability of the non-parametric non-uniform intensity normalization (N3) technique to correct non-uniformity in 72 volumetric brain MR scans from the preparatory phase of the Alzheimer's Disease Neuroimaging Initiative (ADNI). Normal elderly subjects (n=18) were scanned on different 3-T scanners with a multichannel phased array receiver coil at baseline, using magnetization prepared rapid gradient echo (MP-RAGE) and spoiled gradient echo (SPGR) pulse sequences, and again 2 weeks later. When applying N3, we used five brain masks of varying accuracy and four spline smoothing distances (d=50, 100, 150 and 200 mm) to ascertain which combination of parameters optimally reduces the non-uniformity. We used the normalized white matter intensity variance (standard deviation/mean) to ascertain quantitatively the correction for a single scan; we used the variance of the normalized difference image to assess quantitatively the consistency of the correction over time from registered scan pairs. Our results showed statistically significant (p<0.01) improvement in uniformity for individual scans and reduction in the normalized difference image variance when using masks that identified distinct brain tissue classes, and when using smaller spline smoothing distances (e.g., 50-100 mm) for both MP-RAGE and SPGR pulse sequences. These optimized settings may assist future large-scale studies where 3-T scanners and phased array receiver coils are used, such as ADNI, so that intensity non-uniformity does not influence the power of MR imaging to detect disease progression and the factors that influence it.

Brain tissue segmentation based on DTI data

We present a method for automated brain tissue segmentation based on the multi-channel fusion of diffusion tensor imaging (DTI) data. The method is motivated by the evidence that independent tissue segmentation based on DTI parametric images provides complementary information of tissue contrast to the tissue segmentation based on structural MRI data. This has important applications in defining accurate tissue maps when fusing structural data with diffusion data. In the absence of structural data, tissue segmentation based on DTI data provides an alternative means to obtain brain tissue segmentation. Our approach to the tissue segmentation based on DTI data is to classify the brain into two compartments by utilizing the tissue contrast existing in a single channel. Specifically, because the apparent diffusion coefficient (ADC) values in the cerebrospinal fluid (CSF) are more than twice that of gray matter (GM) and white matter (WM), we use ADC images to distinguish CSF and non-CSF tissues. Additionally, fractional anisotropy (FA) images are used to separate WM from non-WM tissues, as highly directional white matter structures have much larger fractional anisotropy values. Moreover, other channels to separate tissue are explored, such as eigenvalues of the tensor, relative anisotropy (RA), and volume ratio (VR). We developed an approach based on the Simultaneous Truth and Performance Level Estimation (STAPLE) algorithm that combines these two-class maps to obtain a complete tissue segmentation map of CSF, GM, and WM. Evaluations are provided to demonstrate the performance of our approach. Experimental results of applying this approach to brain tissue segmentation and deformable registration of DTI data and spoiled gradient-echo (SPGR) data are also provided.

Automated segmentation and classification of high throughput yeast assay spots

Several technologies for characterizing genes and proteins from humans and other organisms use yeast growth or color development as read outs. The yeast two-hybrid assay, for example, detects protein-protein interactions by measuring the growth of yeast on a specific solid medium, or the ability of the yeast to change color when grown on a medium containing a chromogenic substrate. Current systems for analyzing the results of these types of assays rely on subjective and inefficient scoring of growth or color by human experts. Here, an image analysis system is described for scoring yeast growth and color development in high throughput biological assays. The goal is to locate the spots and score them in color images of two types of plates named "X-Gal" and "growth assay" plates, with uniformly placed spots (cell areas) on each plate (both plates in one image). The scoring system relies on color for the X-Gal spots, and texture properties for the growth assay spots. A maximum likelihood projection-based segmentation is developed to automatically locate spots of yeast on each plate. Then color histogram and wavelet texture features are extracted for scoring using an optimal linear transformation. Finally, an artificial neural network is used to score the X-Gal and growth assay spots using the extracted features. The performance of the system is evaluated using spots of 60 images. After training the networks using training and validation sets, the system was assessed on the test set. The overall accuracies of 95.4% and 88.2% are achieved, respectively, for scoring the X-Gal and growth assay spots.

Using the logarithm of odds to define a vector space on probabilistic atlases

The logarithm of the odds ratio (LogOdds) is frequently used in areas such as artificial neural networks, economics, and biology, as an alternative representation of probabilities. Here, we use LogOdds to place probabilistic atlases in a linear vector space. This representation has several useful properties for medical imaging. For example, it not only encodes the shape of multiple anatomical structures but also captures some information concerning uncertainty. We demonstrate that the resulting vector space operations of addition and scalar multiplication have natural probabilistic interpretations. We discuss several examples for placing label maps into the space of LogOdds. First, we relate signed distance maps, a widely used implicit shape representation, to LogOdds and compare it to an alternative that is based on smoothing by spatial Gaussians. We find that the LogOdds approach better preserves shapes in a complex multiple object setting. In the second example, we capture the uncertainty of boundary locations by mapping multiple label maps of the same object into the LogOdds space. Third, we define a framework for non-convex interpolations among atlases that capture different time points in the aging process of a population. We evaluate the accuracy of our representation by generating a deformable shape atlas that captures the variations of anatomical shapes across a population. The deformable atlas is the result of a principal component analysis within the LogOdds space. This atlas is integrated into an existing segmentation approach for MR images. We compare the performance of the resulting implementation in segmenting 20 test cases to a similar approach that uses a more standard shape model that is based on signed distance maps. On this data set, the Bayesian classification model with our new representation outperformed the other approaches in segmenting subcortical structures.

A unified approach for lesion segmentation on MRI of multiple sclerosis

Accurate determination of lesion volumes on brain MR images is hampered by the presence of a large number of false positive and negative classifications. A strategy that combines parametric and nonparametric techniques is developed and implemented for minimizing the false classifications. Initially, CSF and lesions are segmented using Parzen window classifier. Image processing, morphological operations, and ratio map of proton density (PD) and T2 weighted images are used for minimizing false positives. Lesions are delineated using fuzzy connectedness principle. Contextual information was used for minimizing false negative lesion classifications. Gray and white matter classification is realized using HMRF-EM algorithm.

A hierarchical algorithm for MR brain image parcellation

We introduce an algorithm for segmenting brain magnetic resonance (MR) images into anatomical compartments such as the major tissue classes and neuro-anatomical structures of the gray matter. The algorithm is guided by prior information represented within a tree structure. The tree mirrors the hierarchy of anatomical structures and the subtrees correspond to limited segmentation problems. The solution to each problem is estimated via a conventional classifier. Our algorithm can be adapted to a wide range of segmentation problems by modifying the tree structure or replacing the classifier. We evaluate the performance of our new segmentation approach by revisiting a previously published statistical group comparison between first-episode schizophrenia patients, first-episode affective psychosis patients, and comparison subjects. The original study is based on 50 MR volumes in which an expert identified the brain tissue classes as well as the superior temporal gyrus, amygdala, and hippocampus. We generate analogous segmentations using our new method and repeat the statistical group comparison. The results of our analysis are similar to the original findings, except for one structure (the left superior temporal gyrus) in which a trend-level statistical significance (p = 0.07) was observed instead of statistical significance.

Automatic cortical segmentation in the developing brain

The segmentation of neonatal cortex from magnetic resonance (MR) images is much more challenging than the segmentation of cortex in adults. The main reason is the inverted contrast between grey matter (GM) and white matter (WM) that occurs when myelination is incomplete. This causes mislabeled partial volume voxels, especially at the interface between GM and cerebrospinal fluid (CSF). We propose a fully automatic cortical segmentation algorithm, detecting these mislabeled voxels using a knowledge-based approach and correcting errors by adjusting local priors to favor the correct classification. Our results show that the proposed algorithm corrects errors in the segmentation of both GM and WM compared to the classic EM scheme. The segmentation algorithm has been tested on 25 neonates with the gestational ages ranging from approximately 27 to 45 weeks. Quantitative comparison to the manual segmentation demonstrates good performance of the method (mean Dice similarity: 0.758 +/- 0.037 for GM and 0.794 +/- 0.078 for WM).

Automatic segmentation and reconstruction of the cortex from neonatal MRI

Segmentation and reconstruction of cortical surfaces from magnetic resonance (MR) images are more challenging for developing neonates than adults. This is mainly due to the dynamic changes in the contrast between gray matter (GM) and white matter (WM) in both T1- and T2-weighted images (T1w and T2w) during brain maturation. In particular in neonatal T2w images WM typically has higher signal intensity than GM. This causes mislabeled voxels during cortical segmentation, especially in the cortical regions of the brain and in particular at the interface between GM and cerebrospinal fluid (CSF). We propose an automatic segmentation algorithm detecting these mislabeled voxels and correcting errors caused by partial volume effects. Our results show that the proposed algorithm corrects errors in the segmentation of both GM and WM compared to the classic expectation maximization (EM) scheme. Quantitative validation against manual segmentation demonstrates good performance (the mean Dice value: 0.758+/-0.037 for GM and 0.794+/-0.078 for WM). The inner, central and outer cortical surfaces are then reconstructed using implicit surface evolution. A landmark study is performed to verify the accuracy of the reconstructed cortex (the mean surface reconstruction error: 0.73 mm for inner surface and 0.63 mm for the outer). Both segmentation and reconstruction have been tested on 25 neonates with the gestational ages ranging from approximately 27 to 45 weeks. This preliminary analysis confirms previous findings that cortical surface area and curvature increase with age, and that surface area scales to cerebral volume according to a power law, while cortical thickness is not related to age or brain growth.

Multi-spectral brain tissue segmentation using automatically trained k-Nearest-Neighbor classification

Conventional k-Nearest-Neighbor (kNN) classification, which has been successfully applied to classify brain tissue in MR data, requires training on manually labeled subjects. This manual labeling is a laborious and time-consuming procedure. In this work, a new fully automated brain tissue classification procedure is presented, in which kNN training is automated. This is achieved by non-rigidly registering the MR data with a tissue probability atlas to automatically select training samples, followed by a post-processing step to keep the most reliable samples. The accuracy of the new method was compared to rigid registration-based training and to conventional kNN-based segmentation using training on manually labeled subjects for segmenting gray matter (GM), white matter (WM) and cerebrospinal fluid (CSF) in 12 data sets. Furthermore, for all classification methods, the performance was assessed when varying the free parameters. Finally, the robustness of the fully automated procedure was evaluated on 59 subjects. The automated training method using non-rigid registration with a tissue probability atlas was significantly more accurate than rigid registration. For both automated training using non-rigid registration and for the manually trained kNN classifier, the difference with the manual labeling by observers was not significantly larger than inter-observer variability for all tissue types. From the robustness study, it was clear that, given an appropriate brain atlas and optimal parameters, our new fully automated, non-rigid registration-based method gives accurate and robust segmentation results. A similarity index was used for comparison with manually trained kNN. The similarity indices were 0.93, 0.92 and 0.92, for CSF, GM and WM, respectively. It can be concluded that our fully automated method using non-rigid registration may replace manual segmentation, and thus that automated brain tissue segmentation without laborious manual training is feasible.

Volumetric neuroimage analysis extensions for the MIPAV software package

We describe a new collection of publicly available software tools for performing quantitative neuroimage analysis. The tools perform semi-automatic brain extraction, tissue classification, Talairach alignment, and atlas-based measurements within a user-friendly graphical environment. They are implemented as plug-ins for MIPAV, a freely available medical image processing software package from the National Institutes of Health. Because the plug-ins and MIPAV are implemented in Java, both can be utilized on nearly any operating system platform. In addition to the software plug-ins, we have also released a digital version of the Talairach atlas that can be used to perform regional volumetric analyses. Several studies are conducted applying the new tools to simulated and real neuroimaging data sets.