Adaptive segmentation of MRI data

Statistical cerebrovascular segmentation in three-dimensional rotational angiography based on maximum intensity projections

Segmentation of three-dimensional rotational angiography (3D-RA) can provide quantitative 3D morphological information of vasculature. The expectation maximization-(EM-) based segmentation techniques have been widely used in the medical image processing community, because of the implementation simplicity, and computational efficiency of the approach. In a brain 3D-RA, vascular regions usually occupy a very small proportion (around 1%) inside an entire image volume. This severe imbalance between the intensity distributions of vessels and background can lead to inaccurate statistical modeling in the EM-based segmentation methods, and thus adversely affect the segmentation quality for 3D-RA. In this paper we present a new method for the extraction of vasculature in 3D-RA images. The new method is fully automatic and computationally efficient. As compared with the original 3D-RA volume, there is a larger proportion (around 20%) of vessels in its corresponding maximum intensity projection (MIP) image. The proposed method exploits this property to increase the accuracy of statistical modeling with the EM algorithm. The algorithm takes an iterative approach to compiling the 3D vascular segmentation progressively with the segmentation of MIP images along the three principal axes, and use a winner-takes-all strategy to combine the results obtained along individual axes. Experimental results on 12 3D-RA clinical datasets indicate that the segmentations obtained by the new method exhibit a high degree of agreement to the ground truth segmentations and are comparable to those produced by the manual optimal global thresholding method.

76-space analysis of grey matter diffusivity: methods and applications

Diffusion-weighted imaging (DWI) and diffusion tensor imaging (DTI) allow in vivo investigation of molecular motion of tissue water at a microscopic level in cerebral gray matter (GM) and white matter (WM). DWI/DTI measure of water diffusion has been proven to be invaluable for the study of many neurodegenerative diseases (e.g., Alzheimer's disease and Creutzfeldt-Jakob disease) that predominantly involve GM. Thus, quantitative analysis of GM diffusivity is of scientific interest and is promised to have a clinical impact on the investigation of normal brain aging and neuropathology. In this paper, we propose an automated framework for analysis of GM diffusivity in 76 standard anatomic subdivisions of gray matter to facilitate studies of neurodegenerative and other gray matter neurological diseases. The computational framework includes three enabling technologies: (1) automatic parcellation of structural MRI GM into 76 precisely defined neuroanatomic subregions ("76-space"), (2) automated segmentation of GM, WM and CSF based on DTI data, and (3) automatic measurement of the average apparent diffusion coefficient (ADC) in each segmented GM subregion. We evaluate and validate this computational framework for 76-space GM diffusivity analysis using data from normal volunteers and from patients with Creutzfeldt-Jakob disease.

Segmentation of brain MR images using genetically guided clustering

This paper presents a novel algorithm for fuzzy segmentation of magnetic resonance imaging (MRI) data and estimation of intensity inhomogeneities using fuzzy logic. The proposed algorithm is formulated by modifying the objective function of the standard fuzzy c-means (FCM) algorithm to compensate for such inhomogeneities and to allow the labeling of a pixel to be influenced by the labels in its immediate neighborhood. Clustering algorithms such as FCM that use calculus based optimization methods can be trapped by local extrema in the process of optimizing the clustering criterion. They are also very sensitive to initialization. The proposed algorithm uses Genetic Algorithm (GA) to optimize the modified fuzzy c-means function. The performance of the algorithm is evaluated on a series of MR images of the brain.

Data-driven brain MRI segmentation supported on edge confidence and a priori tissue information

Brain magnetic resonance imaging segmentation is accomplished in this work by applying nonparametric density estimation, using the mean shift algorithm in the joint spatial-range domain. The quality of the class boundaries is improved by including an edge confidence map, that represents the confidence of truly being in the presence of a border between adjacent regions; an adjacency graph is then constructed with the labeled regions, and analyzed and pruned to merge adjacent regions. In order to assign image regions to a cerebral tissue type, a spatial normalization between image data and standard probability maps is carried out, so that for each structure a maximum a posteriori probability criterion is applied. The method was applied to synthetic and real images, keeping all parameters constant throughout the process for each type of data. The combination of region segmentation and edge detection proved to be a robust technique, as adequate clusters were automatically identified, regardless of the noise level and bias. In a comparison with reference segmentations, average Tanimoto indexes of 0.90-0.99 were obtained for synthetic data and of 0.59-0.99 for real data, considering gray matter, white matter, and background.

Spline-based probabilistic model for anatomical landmark detection

In medical imaging, finding landmarks that provide biologically meaningful correspondences is often a challenging and time-consuming manual task. In this paper we propose a generic and simple algorithm for landmarking non-cortical brain structures automatically. We use a probabilistic model of the image intensities based on the deformation of a tissue probability map, learned from a training set of hand-landmarked images. In this setting, estimating the location of the landmarks in a new image is equivalent to finding, by likelihood maximization, the "best" deformation from the tissue probability map to the image. The resulting algorithm is able to handle arbitrary types and numbers of landmarks. We demonstrate our algorithm on the detection of 3 landmarks of the hippocampus in brain MR images.

Segmentation of Brain MRI in Young Children

This paper describes an automatic tissue segmentation algorithm for brain MRI of young children. Existing segmentation methods developed for the adult brain do not take into account the specific tissue properties present in the brain MRI of young children. We examine the suitability of state-of-the-art methods developed for the adult brain when applied to the segmentation of the young child brain MRI. We develop a method of creation of a population-specific atlas from young children using a single manual segmentation. The method is based on non-linear propagation of the segmentation into population and subsequent affine alignment into a reference space and averaging. Using this approach we significantly improve the performance of the popular EM segmentation algorithm on brain MRI of young children.

A fuzzy c-means (FCM)-based approach for computerized segmentation of breast lesions in dynamic contrast-enhanced MR images

RATIONALE AND OBJECTIVES

Accurate quantification of the shape and extent of breast tumors has a vital role in nearly all applications of breast magnetic resonance (MR) imaging (MRI). Specifically, tumor segmentation is a key component in the computerized assessment of likelihood of malignancy. However, manual delineation of lesions in four-dimensional MR images is labor intensive and subject to interobserver and intraobserver variations. We developed a computerized lesion segmentation method that has the advantage of being automatic, efficient, and objective.

MATERIALS AND METHODS

We present a fuzzy c-means (FCM) clustering-based method for the segmentation of breast lesions in three dimensions from contrast-enhanced MR images. The proposed lesion segmentation algorithm consists of six consecutive stages: region of interest (ROI) selection by a human operator, lesion enhancement within the selected ROI, application of FCM on the enhanced ROI, binarization of the lesion membership map, connected-component labeling and object selection, and hole-filling on the selected object. We applied the algorithm to a clinical MR database consisting of 121 primary mass lesions. Manual segmentation of the lesions by an expert MR radiologist served as a reference in the evaluation of the computerized segmentation method. We also compared the proposed algorithm with a previously developed volume-growing (VG) method.

RESULTS

For the 121 mass lesions in our database, 97% of lesions were segmented correctly by means of the proposed FCM-based method at an overlap threshold of 0.4, whereas 84% of lesions were correctly segmented by means of the VG method.

CONCLUSION

Our proposed algorithm for breast-lesion segmentation in dynamic contrast-enhanced MRI was shown to be effective and efficient.

A Fast and Automatic Method to Correct Intensity Inhomogeneity in MR Brain Images

This paper presents a method to improve the semi-automatic method for intensity inhomogeneity correction by Dawant et al. through introducing a fully automatic approach to reference points generation, which is based on order statistics and integrates information from the fine to coarse scale representations of the input image. The method has been validated and compared with two popular methods, N3 and BFC. Advantages of the proposed method are demonstrated.

Classification of spatiotemporal hemodynamics from brain perfusion MR images using expectation-maximization estimation with finite mixture of multivariate gaussian distributions

The ability to cluster different perfusion compartments in the brain is critical for analyzing brain perfusion. This study presents a method based on a mixture of multivariate Gaussians (MoMG) and the expectation-maximization (EM) algorithm to dissect various perfusion compartments from dynamic susceptibility contrast (DSC) MR images so that each compartment comprises pixels of similar signal-time curves. This EM-based method provides an objective way to 1) delineate an area to serve as the in-plane arterial input function (AIF) of the feeding artery for adjacent tissues to better quantify the relative cerebral blood volume (rCBV), relative cerebral blood flow (rCBF), and mean transit time (MTT); 2) demarcate regions with abnormal perfusion derangement to facilitate diagnosis; and 3) obtain parametric maps with supplementary information, such as temporal scenarios and recirculation of contrast agent. Results from normal subjects show that perfusion cascade manifests (in order of appearance) the arteries, gray matter (GM), white matter (WM), veins and sinuses, and choroid plexus mixed with cerebrospinal fluid (CSF). The averaged rCBV, rCBF, and MTT ratios between GM and WM are in good agreement with those in the literature. Results from a patient with cerebral arteriovenous malformation (CAVM) showed distinct spatiotemporal characteristics between perfusion patterns, which allowed differentiation between pathological and nonpathological areas.

Segmenting magnetic resonance images via hierarchical mixture modelling

We present a statistically innovative as well as scientifically and practically relevant method for automatically segmenting magnetic resonance images using hierarchical mixture models. Our method is a general tool for automated cortical analysis which promises to contribute substantially to the science of neuropsychiatry. We demonstrate that our method has advantages over competing approaches on a magnetic resonance brain imagery segmentation task.

Morphological exponential entropy driven-HUM

This paper presents an improvement to the exponential entropy driven-homomorphic unsharp masking (E(2)D-HUM) algorithm devoted to illumination artifact suppression on magnetic resonance images. E(2)D-HUM requires a segmentation step to remove dark regions in the foreground whose intensity is comparable with background, because strong edges produce streak artifacts on the tissues. This new version of the algorithm keeps the same good properties of E(2)D-HUM without a segmentation phase, whose parameters should be chosen in relation to the image.

Automatic segmentation of MR images of the developing newborn brain

This paper describes an automatic tissue segmentation method for newborn brains from magnetic resonance images (MRI). The analysis and study of newborn brain MRI is of great interest due to its potential for studying early growth patterns and morphological changes in neurodevelopmental disorders. Automatic segmentation of newborn MRI is a challenging task mainly due to the low intensity contrast and the growth process of the white matter tissue. Newborn white matter tissue undergoes a rapid myelination process, where the nerves are covered in myelin sheathes. It is necessary to identify the white matter tissue as myelinated or non-myelinated regions. The degree of myelination is a fractional voxel property that represents regional changes of white matter as a function of age. Our method makes use of a registered probabilistic brain atlas. The method first uses robust graph clustering and parameter estimation to find the initial intensity distributions. The distribution estimates are then used together with the spatial priors to perform bias correction. Finally, the method refines the segmentation using training sample pruning and non-parametric kernel density estimation. Our results demonstrate that the method is able to segment the brain tissue and identify myelinated and non-myelinated white matter regions.

An EM algorithm for shape classification based on level sets

In this paper, we propose an expectation-maximization (EM) approach to separate a shape database into different shape classes, while simultaneously estimating the shape contours that best exemplify each of the different shape classes. We begin our formulation by employing the level set function as the shape descriptor. Next, for each shape class we assume that there exists an unknown underlying level set function whose zero level set describes the contour that best represents the shapes within that shape class. The level set function for each example shape in the database is modeled as a noisy measurement of the appropriate shape class's unknown underlying level set function. Based on this measurement model and the judicious introduction of the class labels as the hidden data, our EM formulation calculates the labels for shape classification and estimates the shape contours that best typify the different shape classes. This resulting iterative algorithm is computationally efficient, simple, and accurate. We demonstrate the utility and performance of this algorithm by applying it to two medical applications.

An introduction to MR imaging-based stroke morphometry

The anatomic description of the stroke lesion is an essential component of clinical diagnosis and treatment and has become an established tool in investigations into underlying stroke pathophysiology. Magnetic resonance (MR) imaging permits quantitative evaluation of the distributed consequences of the pathologic stroke insult. General properties of stroke effects have emerged using these tools. This article surveys the classes of morphometric data that are available from conventional MR images, the methods for extracting quantitative results, and samples of the application of these methods to stroke. These samples highlight anatomic-based considerations regarding the nature of stroke and its repercussions within the brain parenchyma.

Comparison and validation of tissue modelization and statistical classification methods in T1-weighted MR brain images

This paper presents a validation study on statistical nonsupervised brain tissue classification techniques in magnetic resonance (MR) images. Several image models assuming different hypotheses regarding the intensity distribution model, the spatial model and the number of classes are assessed. The methods are tested on simulated data for which the classification ground truth is known. Different noise and intensity nonuniformities are added to simulate real imaging conditions. No enhancement of the image quality is considered either before or during the classification process. This way, the accuracy of the methods and their robustness against image artifacts are tested. Classification is also performed on real data where a quantitative validation compares the methods' results with an estimated ground truth from manual segmentations by experts. Validity of the various classification methods in the labeling of the image as well as in the tissue volume is estimated with different local and global measures. Results demonstrate that methods relying on both intensity and spatial information are more robust to noise and field inhomogeneities. We also demonstrate that partial volume is not perfectly modeled, even though methods that account for mixture classes outperform methods that only consider pure Gaussian classes. Finally, we show that simulated data results can also be extended to real data.

Intensity non-uniformity correction in MRI: existing methods and their validation

Magnetic resonance imaging is a popular and powerful non-invasive imaging technique. Automated analysis has become mandatory to efficiently cope with the large amount of data generated using this modality. However, several artifacts, such as intensity non-uniformity, can degrade the quality of acquired data. Intensity non-uniformity consists in anatomically irrelevant intensity variation throughout data. It can be induced by the choice of the radio-frequency coil, the acquisition pulse sequence and by the nature and geometry of the sample itself. Numerous methods have been proposed to correct this artifact. In this paper, we propose an overview of existing methods. We first sort them according to their location in the acquisition/processing pipeline. Sorting is then refined based on the assumptions those methods rely on. Next, we present the validation protocols used to evaluate these different correction schemes both from a qualitative and a quantitative point of view. Finally, availability and usability of the presented methods is discussed.

Inhomogeneity correction for brain magnetic resonance images by rank leveling

OBJECTIVE

A postprocessing method of rank filtering inhomogeneity correction using nonlinear rank filtering of magnetic resonance imaging (MRI) scans is described. The method addresses some of the problems of homomorphic unsharp masking (HUM) using mean or median filtering.

METHODS

Maximum rank filtering was used to estimate the bias image, which was then smoothed and used to normalize the original image. The coefficient of variation within and between tissue classes before and after inhomogeneity correction was calculated in simulated brain phantom images and clinical T1-weighted MRI images. Comparison was made with mean filter-based and median filter-based HUM.

RESULTS

Maximum rank filtering reduced within and between class coefficients of variation. Performance of median filtering was inferior to that of mean filtering, and both were inferior to performance of maximum rank filtering.

CONCLUSION

The method is easy to implement and is effective against different bias types. It is less prone to edge effects than mean and median filtering.

Partial volume segmentation of brain magnetic resonance images based on maximum a posteriori probability

Noise, partial volume (PV) effect, and image-intensity inhomogeneity render a challenging task for segmentation of brain magnetic resonance (MR) images. Most of the current MR image segmentation methods focus on only one or two of the above-mentioned effects. The objective of this paper is to propose a unified framework, based on the maximum a posteriori probability principle, by taking all these effects into account simultaneously in order to improve image segmentation performance. Instead of labeling each image voxel with a unique tissue type, the percentage of each voxel belonging to different tissues, which we call a mixture, is considered to address the PV effect. A Markov random field model is used to describe the noise effect by considering the nearby spatial information of the tissue mixture. The inhomogeneity effect is modeled as a bias field characterized by a zero mean Gaussian prior probability. The well-known fuzzy C-mean model is extended to define the likelihood function of the observed image. This framework reduces theoretically, under some assumptions, to the adaptive fuzzy C-mean (AFCM) algorithm proposed by Pham and Prince. Digital phantom and real clinical MR images were used to test the proposed framework. Improved performance over the AFCM algorithm was observed in a clinical environment where the inhomogeneity, noise level, and PV effect are commonly encountered.

Probabilistic segmentation of brain tissue in MR imaging

A new method has been developed for probabilistic segmentation of five different types of brain structures: white matter, gray matter, cerebro-spinal fluid without ventricles, ventricles and white matter lesion in cranial MR imaging. The algorithm is based on information from T1-weighted (T1-w), inversion recovery (IR), proton density-weighted (PD), T2-weighted (T2-w) and fluid attenuation inversion recovery (FLAIR) scans. It uses the K-Nearest Neighbor classification technique that builds a feature space from spatial information and voxel intensities. The technique generates for each tissue type an image representing the probability per voxel being part of it. By application of thresholds on these probability maps, binary segmentations can be obtained. A similarity index (SI) and a probabilistic SI (PSI) were calculated for quantitative evaluation of the results. The influence of each image type on the performance was investigated by alternately leaving out one of the five scan types. This procedure showed that the incorporation of the T1-w, PD or T2-w did not significantly improve the segmentation results. Further investigation indicated that the combination of IR and FLAIR was optimal for segmentation of the five brain tissue types. Evaluation with respect to the gold standard showed that the SI-values for all tissues exceeded 0.8 and all PSI-values exceeded 0.7, implying an excellent agreement.

Shape Based Segmentation of Anatomical Structures in Magnetic Resonance Images

Standard image based segmentation approaches perform poorly when there is little or no contrast along boundaries of different regions. In such cases, segmentation is largely performed manually using prior knowledge of the shape and relative location of the underlying structures combined with partially discernible boundaries. We present an automated approach guided by covariant shape deformations of neighboring structures, which is an additional source of prior information. Captured by a shape atlas, these deformations are transformed into a statistical model using the logistic function. Structure boundaries, anatomical labels, and image inhomogeneities are estimated simultaneously within an Expectation-Maximization formulation of the maximum a posteriori probability estimation problem. We demonstrate the approach on 20 brain magnetic resonance images showing superior performance, particularly in cases where purely image based methods fail.

MRI Fuzzy Segmentation of Brain Tissue Using Neighborhood Attraction With Neural-Network Optimization

Image segmentation is an indispensable process in the visualization of human tissues, particularly during clinical analysis of magnetic resonance (MR) images. Unfortunately, MR images always contain a significant amount of noise caused by operator performance, equipment, and the environment, which can lead to serious inaccuracies with segmentation. A robust segmentation technique based on an extension to the traditional fuzzy c-means (FCM) clustering algorithm is proposed in this paper. A neighborhood attraction, which is dependent on the relative location and features of neighboring pixels, is shown to improve the segmentation performance dramatically. The degree of attraction is optimized by a neural-network model. Simulated and real brain MR images with different noise levels are segmented to demonstrate the superiority of the proposed technique compared to other FCM-based methods. This segmentation method is a key component of an MR image-based classification system for brain tumors, currently being developed. Index Terms-Improved fuzzy c-means clustering (IFCM), magnetic resonance imaging (MRI), neighborhood attraction, segmentation.

Using automated morphometry to detect associations between ERP latency and structural brain MRI in normal adults

Despite the clinical significance of event-related potential (ERP) latency abnormalities, little attention has focused on the anatomic substrate of latency variability. Volume conduction models do not identify the anatomy responsible for delayed neural transmission between neural sources. To explore the anatomic substrate of ERP latency variability in normal adults using automated measures derived from magnetic resonance imaging (MRI), ERPs were recorded in the visual three-stimulus oddball task in 59 healthy participants. Latencies of the P3a and P3b components were measured at the vertex. Measures of local anatomic size in the brain were estimated from structural MRI, using tissue segmentation and deformation morphometry. A general linear model was fitted relating latency to measures of local anatomic size, covarying for intracranial vault volume. Longer P3b latencies were related to contractions in thalamus extending superiorly into the corpus callosum, white matter (WM) anterior to the central sulcus on the left and right, left temporal WM, the right anterior limb of the internal capsule extending into the lenticular nucleus, and larger cerebrospinal fluid volumes. There was no evidence for a relationship between gray matter (GM) volumes and P3b latency. Longer P3a latencies were related to contractions in left temporal WM, and left parietal GM and WM near the interhemispheric fissure. P3b latency variability is related chiefly to WM, thalamus, and lenticular nucleus, whereas P3a latency variability is not related as strongly to anatomy. These results imply that the WM connectivity between generators influences P3b latency more than the generators themselves do.

Tagged magnetic resonance imaging of the heart: a survey

Magnetic resonance imaging (MRI) of the heart with magnetization tagging provides a potentially useful new way to assess cardiac mechanical function, through revealing the local motion of otherwise indistinguishable portions of the heart wall. While still an evolving area, tagged cardiac MRI is already able to provide novel quantitative information on cardiac function. Exploiting this potential requires developing tailored methods for both imaging and image analysis. In this paper, we review some of the progress that has been made in developing such methods for tagged cardiac MRI, as well as some of the ways these methods have been applied to the study of cardiac function.

Correction of bias field in MR images using singularity function analysis

A new approach for correcting bias field in magnetic resonance (MR) images is proposed using the mathematical model of singularity function analysis (SFA), which represents a discrete signal or its spectrum as a weighted sum of singularity functions. Through this model, an MR image's low spatial frequency components corrupted by a smoothly varying bias field are first removed, and then reconstructed from its higher spatial frequency components not polluted by bias field. The thus reconstructed image is then used to estimate bias field for final image correction. The approach does not rely on the assumption that anatomical information in MR images occurs at higher spatial frequencies than bias field. The performance of this approach is evaluated using both simulated and real clinical MR images.

Automated segmentation of neuroanatomical structures in multispectral MR microscopy of the mouse brain

We present the automated segmentation of magnetic resonance microscopy (MRM) images of the C57BL/6J mouse brain into 21 neuroanatomical structures, including the ventricular system, corpus callosum, hippocampus, caudate putamen, inferior colliculus, internal capsule, globus pallidus, and substantia nigra. The segmentation algorithm operates on multispectral, three-dimensional (3D) MR data acquired at 90-microm isotropic resolution. Probabilistic information used in the segmentation is extracted from training datasets of T2-weighted, proton density-weighted, and diffusion-weighted acquisitions. Spatial information is employed in the form of prior probabilities of occurrence of a structure at a location (location priors) and the pairwise probabilities between structures (contextual priors). Validation using standard morphometry indices shows good consistency between automatically segmented and manually traced data. Results achieved in the mouse brain are comparable with those achieved in human brain studies using similar techniques. The segmentation algorithm shows excellent potential for routine morphological phenotyping of mouse models.

Increasing the power of functional maps of the medial temporal lobe by using large deformation diffeomorphic metric mapping

The functional magnetic resonance imagery responses of declarative memory tasks in the medial temporal lobe (MTL) are examined by using large deformation diffeomorphic metric mapping (LDDMM) to remove anatomical variations across subjects. LDDMM is used to map the structures of the MTL in multiple subjects into extrinsic atlas coordinates; these same diffeomorphic mappings are used to transfer the corresponding functional data activation to the same extrinsic coordinates. The statistical power in the averaged LDDMM mapped signals is significantly increased over conventional Talairach-Tournoux averaging. Activation patterns are highly localized within the MTL. Whereas the present demonstration has been aimed at enhancing alignment within the MTL, this technique is general and can be applied throughout the brain.

A stochastic model for studying the laminar structure of cortex from MRI

The human cerebral cortex is a laminar structure about 3 mm thick, and is easily visualized with current magnetic resonance (MR) technology. The thickness of the cortex varies locally by region, and is likely to be influenced by such factors as development, disease and aging. Thus, accurate measurements of local cortical thickness are likely to be of interest to other researchers. We develop a parametric stochastic model relating the laminar structure of local regions of the cerebral cortex to MR image data. Parameters of the model include local thickness, and statistics describing white, gray and cerebrospinal fluid (CSF) image intensity values as a function of the normal distance from the center of a voxel to a local coordinate system anchored at the gray/white matter interface. Our fundamental data object, the intensity-distance histogram (IDH), is a two-dimensional (2-D) generalization of the conventional 1-D image intensity histogram, which indexes voxels not only by their intensity value, but also by their normal distance to the gray/white interface. We model the IDH empirically as a marked Poisson process with marking process a Gaussian random field model of image intensity indexed against normal distance. In this paper, we relate the parameters of the IDH model to the local geometry of the cortex. A maximum-likelihood framework estimates the parameters of the model from the data. Here, we show estimates of these parameters for 10 volumes in the posterior cingulate, and 6 volumes in the anterior and posterior banks of the central sulcus. The accuracy of the estimates is quantified via Cramer-Rao bounds. We believe that this relatively crude model can be extended in a straightforward fashion to other biologically and theoretically interesting problems such as segmentation, surface area estimation, and estimating the thickness distribution in a variety of biologically relevant contexts.

Unified segmentation

A probabilistic framework is presented that enables image registration, tissue classification, and bias correction to be combined within the same generative model. A derivation of a log-likelihood objective function for the unified model is provided. The model is based on a mixture of Gaussians and is extended to incorporate a smooth intensity variation and nonlinear registration with tissue probability maps. A strategy for optimising the model parameters is described, along with the requisite partial derivatives of the objective function.

ANATOMICAL GUIDED SEGMENTATION WITH NON-STATIONARY TISSUE CLASS DISTRIBUTIONS IN AN EXPECTATION-MAXIMIZATION FRAMEWORK

High quality segmentation of brain MR images is a challenging task. To deal with this problem many automatic segmentation methods rely on atlas information of anatomical structures. We further investigate this line of research by introducing hierarchical representations of anatomical structures in an Expectation-Maximization framework. This new approach enables us to divide a complex segmentation scenario into less difficult sub-problems reducing the scenario's statistical complexity. We demonstrate the method's strength by segmenting a set of brain MR images into 31 different anatomical structures as well as comparing it to other methods.

Chronic active heavy drinking and family history of problem drinking modulate regional brain tissue volumes

The goals of this study were to measure if chronic active heavy drinking is associated with brain volume loss in non-treatment seeking men and women, and to assess the effect of positive family history of problem drinking on brain structure in heavy drinkers. Automated image processing was used to analyze high-resolution T1-weighted magnetic resonance images from 49 active heavy drinkers and 49 age- and sex-matched light drinkers, yielding gray matter, white matter and cerebrospinal fluid (CSF) volumes within the frontal, temporal, parietal and occipital lobes. Regional brain volume measures were compared as a function of group, sex and their interaction. Within heavy drinkers, volumes were correlated with measures of alcohol consumption and compared as a function of family history of problem drinking. Deformation morphometry explored localized patterns of atrophy associated with heavy drinking or severity of drinking. We found significant gray matter volume losses, but no white matter losses, in active heavy drinkers compared with light drinkers. Women had greater gray matter and smaller white matter and CSF volumes as a percentage of intracranial vault than men. Within heavy drinkers, smaller gray matter volumes were associated with higher current levels of drinking and older age, while a positive family history of problem drinking was associated with smaller CSF volumes. Community-dwelling heavy drinkers who are not in alcoholism treatment have dose-related gray matter volume losses, and family history of problem drinking ameliorates some structural consequences of heavy drinking.

MRI intensity inhomogeneity correction by combining intensity and spatial information

We propose a novel fully automated method for retrospective correction of intensity inhomogeneity, which is an undesired phenomenon in many automatic image analysis tasks, especially if quantitative analysis is the final goal. Besides most commonly used intensity features, additional spatial image features are incorporated to improve inhomogeneity correction and to make it more dynamic, so that local intensity variations can be corrected more efficiently. The proposed method is a four-step iterative procedure in which a non-parametric inhomogeneity correction is conducted. First, the probability distribution of image intensities and corresponding second derivatives is obtained. Second, intensity correction forces, condensing the probability distribution along the intensity feature, are computed for each voxel. Third, the inhomogeneity correction field is estimated by regularization of all voxel forces, and fourth, the corresponding partial inhomogeneity correction is performed. The degree of inhomogeneity correction dynamics is determined by the size of regularization kernel. The method was qualitatively and quantitatively evaluated on simulated and real MR brain images. The obtained results show that the proposed method does not corrupt inhomogeneity-free images and successfully corrects intensity inhomogeneity artefacts even if these are more dynamic.

Has your patient's multiple sclerosis lesion burden or brain atrophy actually changed?

Changes in mean magnetic resonance imaging (MRI)-derived measurements between patient groups are often used to determine outcomes in therapeutic trials and other longitudinal studies of multiple sclerosis (MS). However, in day-to-day clinical practice the changes within individual patients may also be of interest In this paper, we estimated the measurement error of an automated brain tissue quantification algorithm and determined the thresholds for statistically significant change of MRI-derived T2 lesion volume and brain atrophy in individual patients. Twenty patients with MS were scanned twice within 30 min. Brain tissue volumes were measured using the computer algorithm. Brain atrophy was estimated by calculation of brain parenchymal fraction. The threshold of change between repeated scans that represented statistically significant change beyond measurement error with 95% certainty was 0.65 mL for T2 lesion burden and 0.0056 for brain parenchymal fraction. Changes in lesion burden and brain atrophy below these thresholds can be safely (with 95% certainty) explained by measurement variability alone. These values provide clinical neurologists with a useful reference to interpret MRI-derived measures in individual patients.

Computational anatomy: shape, growth, and atrophy comparison via diffeomorphisms

Computational anatomy (CA) is the mathematical study of anatomy I in I = I(alpha) o G, an orbit under groups of diffeomorphisms (i.e., smooth invertible mappings) g in G of anatomical exemplars I(alpha) in I. The observable images are the output of medical imaging devices. There are three components that CA examines: (i) constructions of the anatomical submanifolds, (ii) comparison of the anatomical manifolds via estimation of the underlying diffeomorphisms g in G defining the shape or geometry of the anatomical manifolds, and (iii) generation of probability laws of anatomical variation P(.) on the images I for inference and disease testing within anatomical models. This paper reviews recent advances in these three areas applied to shape, growth, and atrophy.

Two methods for validating brain tissue classifiers

In this paper, we present an evaluation of seven automatic brain tissue classifiers based on level of agreements. A number of agreement measures are explained, and we show how they can be used to compare different segmentation techniques. We use the Simultaneous Truth and Performance Level Estimation (STAPLE) of Warfield et al. but also introduce a novel evaluation technique based on the Williams' index. The methods are evaluated using these two techniques on a population of forty subjects, each having an SPGR scan and a co-registered T2 weighted scan. We provide an interpretation of the results and show how similar the output of the STAPLE analysis and Williams' index are. When no ground truth is required, we recommend the use of Williams' index as it is easy and fast to compute.

A unifying approach to registration, segmentation, and intensity correction

We present a statistical framework that combines the registration of an atlas with the segmentation of magnetic resonance images. We use an Expectation Maximization-based algorithm to find a solution within the model, which simultaneously estimates image inhomogeneities, anatomical labelmap, and a mapping from the atlas to the image space. An example of the approach is given for a brain structure-dependent affine mapping approach. The algorithm produces high quality segmentations for brain tissues as well as their substructures. We demonstrate the approach on a set of 22 magnetic resonance images. In addition, we show that the approach performs better than similar methods which separate the registration and segmentation problems.

Many heads are better than one: jointly removing bias from multiple MRIs using nonparametric maximum likelihood

The correction of multiplicative bias in magnetic resonance images is an important problem in medical image processing, especially as a preprocessing step for quantitative measurements and other numerical procedures. Most previous approaches have used a maximum likelihood method to increase the probability of the pixels in a single image by adaptively estimating a correction to the unknown image bias field. The pixel probabilities are defined either in terms of a pre-existing tissue model, or nonparametrically in terms of the image's own pixel values. In both cases, the specific location of a pixel in the image does not influence the probability calculation. Our approach, similar to methods of joint registration, simultaneously eliminates the bias from a set of images of the same anatomy, but from different patients. We use the statistics from the same location across different patients' images, rather than within an image, to eliminate bias fields from all of the images simultaneously. Evaluating the likelihood of a particular voxel in one patient's scan with respect to voxels in the same location in a set of other patients' scans disambiguates effects that might be due to either bias fields or anatomy. We present a variety of "two-dimensional" experimental results (working with one image from each patient) showing how our method overcomes serious problems experienced by other methods. We also present preliminary results on full three-dimensional volume correction across patients.

An in vivo MRI study of prefrontal cortical complexity in first-episode psychosis

OBJECTIVE

The purpose of this study was to investigate abnormalities in the surface complexity of the prefrontal cortex and in the hemispheric asymmetry of cortical complexity in first-episode patients with schizophrenia.

METHOD

An estimate of the surface complexity of the prefrontal cortex was derived from the number of voxels along the boundary between gray matter and CSF. Magnetic resonance imaging scans were acquired from patients with a first episode of schizophrenia (N=17), patients with a first episode of affective psychosis (N=17), and normal comparison subjects (N=17), age-matched within a narrow age range (18-29 years). This study group was the focus of a previous study that showed lower prefrontal cortical volume in patients with schizophrenia.

RESULTS

Prefrontal cortical complexity was not significantly different among the groups. However, the schizophrenia patients differed significantly from the normal comparison subjects in asymmetry, with the schizophrenia patients showing less left-greater-than-right asymmetry in cortical complexity than the comparison subjects.

CONCLUSIONS

An abnormal pattern of asymmetry in the prefrontal cortex of first-episode patients with schizophrenia provides evidence for a neurodevelopmental mechanism in the etiology of schizophrenia.

Evaluation of accuracy in MS lesion volumetry using realistic lesion phantoms

RATIONALE AND OBJECTIVES

Quantitative analysis of such small structures as focal lesions in patients with multiple sclerosis (MS) is an important issue in both diagnosis and therapy monitoring. To reach clinical relevance, the reproducibility and accuracy of a proposed method have to be validated. We propose a framework for the generation of realistic digital phantoms of MS lesions of known volumes and their incorporation into a magnetic resonance (MR) data set of a healthy volunteer.

MATERIALS AND METHODS

We generated 54 data sets from a multispectral brain scan of a healthy volunteer with incorporated MS lesion phantoms. Lesion phantoms were created using different shapes (three), sizes (six), and orientations (three). An evaluation is carried out from a manual analysis of three human experts and two different semiautomatic approaches, with and without explicit modeling of partial volume effects (PVEs).

RESULTS

Intraobserver and interobserver studies were performed for the phantom data sets. All experts overestimated the true lesion volume for any phantom data set (median overestimation between 42.9% and 63.2%). Relative error and variability increased with decreasing lesion size. Similar results were obtained for the semiautomatic approach without PVE modeling. Only the approach with explicit PVE modeling was capable of generating accurate volumetric results with low systematic error.

CONCLUSION

The proposed framework based on realistic lesion phantoms incorporated into an MR scan allows for quantitative assessment of the accuracy of manual and automated lesion volumetry. Results clearly show the importance of an improved gold standard in lesion volumetry beyond voxel counting.

Fast and robust parameter estimation for statistical partial volume models in brain MRI

Due to the finite spatial resolution of imaging devices, a single voxel in a medical image may be composed of mixture of tissue types, an effect known as partial volume effect (PVE). Partial volume estimation, that is, the estimation of the amount of each tissue type within each voxel, has received considerable interest in recent years. Much of this work has been focused on the mixel model, a statistical model of PVE. We propose a novel trimmed minimum covariance determinant (TMCD) method for the estimation of the parameters of the mixel PVE model. In this method, each voxel is first labeled according to the most dominant tissue type. Voxels that are prone to PVE are removed from this labeled set, following which robust location estimators with high breakdown points are used to estimate the mean and the covariance of each tissue class. Comparisons between different methods for parameter estimation based on classified images as well as expectation--maximization-like (EM-like) procedure for simultaneous parameter and partial volume estimation are reported. The robust estimators based on a pruned classification as presented here are shown to perform well even if the initial classification is of poor quality. The results obtained are comparable to those obtained using the EM-like procedure, but require considerably less computation time. Segmentation results of real data based on partial volume estimation are also reported. In addition to considering the parameter estimation problem, we discuss differences between different approximations to the complete mixel model. In summary, the proposed TMCD method allows for the accurate, robust, and efficient estimation of partial volume model parameters, which is crucial to a variety of brain MRI data analysis procedures such as the accurate estimation of tissue volumes and the accurate delineation of the cortical surface.

Correction of differential intensity inhomogeneity in longitudinal MR images

Longitudinal MR imaging is increasingly being used to measure cerebral atrophy progression in dementia and other neurological disorders. Differences in intensity inhomogeneity between serial scans can confound these measurements. This differential bias also distorts nonlinear registration and makes both manual and automated segmentation of tissue type less reliable. A technique is described for the correction of this differential bias that makes no assumptions about signal distribution, bias field or signal homogeneity. Instead, the bias field calculation is performed on the basis that the remaining structure in the difference image of registered serial scans has small-scale structure. The differential bias field is of much larger scale and can thus be obtained by applying an appropriate filter to the difference image. The serial scan pair is then corrected for the differential bias field and atrophy measurement can be performed on the corrected scan pair. Application of a known, simulated bias field to real serial MR images was shown to alter atrophy measurements significantly. The differential correction method recovered the applied differential bias field and thereby improved atrophy measurements. This method was then applied to serial imaging in patients with dementia using a set of serial scan pairs with visually identified, significant differential bias and a set of scan pairs with negligible differential bias. Differential bias correction specifically reduced the variance of the atrophy measure significantly for the scans with significant differential bias.

Validation of semiautomated methods for quantifying cingulate cortical metrics in schizophrenia

This paper validates semiautomated methods for reconstructing cortical surfaces of the cingulate gyrus from high-resolution magnetic resonance (MR) images. Bayesian segmentation was used to delineate the image voxels into five tissue types: cerebrospinal fluid (CSF), gray matter (GM), white matter (WM), and partial volumes of CSF/GM and GM/WM; the tissues were then recalibrated as CSF, GM, and WM via the Neyman-Pearson Likelihood Ratio Test. To generate cortical surfaces at the interface of GM and WM, the thresholds between the tissue types were first used to reassign partial volume voxels to CSF, GM, and WM with minimum error (that varied from 0.06 to 0.15 for the 10 subjects). Next, topology-correct cortical surfaces were generated and validated with almost all surface vertices lying within one voxel (0.5 mm) of hand contours. Dynamic programming was used to delineate and extract the cingulate gyrus from the cortical surfaces based on its gyral and sulcal boundaries. The intraclass correlation coefficient for surface area obtained by two raters for all 10 surfaces was 0.82. In addition, by repeating the entire procedure three times in one subject, we obtained a coefficient of variation of 0.0438 for surface area.

A hybrid approach to the skull stripping problem in MRI

We present a novel skull-stripping algorithm based on a hybrid approach that combines watershed algorithms and deformable surface models. Our method takes advantage of the robustness of the former as well as the surface information available to the latter. The algorithm first localizes a single white matter voxel in a T1-weighted MRI image, and uses it to create a global minimum in the white matter before applying a watershed algorithm with a preflooding height. The watershed algorithm builds an initial estimate of the brain volume based on the three-dimensional connectivity of the white matter. This first step is robust, and performs well in the presence of intensity nonuniformities and noise, but may erode parts of the cortex that abut bright nonbrain structures such as the eye sockets, or may remove parts of the cerebellum. To correct these inaccuracies, a surface deformation process fits a smooth surface to the masked volume, allowing the incorporation of geometric constraints into the skull-stripping procedure. A statistical atlas, generated from a set of accurately segmented brains, is used to validate and potentially correct the segmentation, and the MRI intensity values are locally re-estimated at the boundary of the brain. Finally, a high-resolution surface deformation is performed that accurately matches the outer boundary of the brain, resulting in a robust and automated procedure. Studies by our group and others outperform other publicly available skull-stripping tools.

Putting our heads together: a consensus approach to brain/non-brain segmentation in T1-weighted MR volumes

We describe an approach to brain extraction from T1-weighted MR volumes that uses a hierarchy of masks created by different models to form a consensus mask. The algorithm (McStrip) incorporates atlas-based extraction via nonlinear warping, intensity-threshold masking with connectivity constraints, and edge-based masking with morphological operations. Volume and boundary metrics were computed to evaluate the reproducibility and accuracy of McStrip against manual brain extraction on 38 scans from normal and ataxic subjects. McStrip masks were reproducible across six repeat scans of a normal subject and were significantly more accurate than the masks produced by any of the individual algorithmic components.

A novel kernelized fuzzy C-means algorithm with application in medical image segmentation

Image segmentation plays a crucial role in many medical imaging applications. In this paper, we present a novel algorithm for fuzzy segmentation of magnetic resonance imaging (MRI) data. The algorithm is realized by modifying the objective function in the conventional fuzzy C-means (FCM) algorithm using a kernel-induced distance metric and a spatial penalty on the membership functions. Firstly, the original Euclidean distance in the FCM is replaced by a kernel-induced distance, and thus the corresponding algorithm is derived and called as the kernelized fuzzy C-means (KFCM) algorithm, which is shown to be more robust than FCM. Then a spatial penalty is added to the objective function in KFCM to compensate for the intensity inhomogeneities of MR image and to allow the labeling of a pixel to be influenced by its neighbors in the image. The penalty term acts as a regularizer and has a coefficient ranging from zero to one. Experimental results on both synthetic and real MR images show that the proposed algorithms have better performance when noise and other artifacts are present than the standard algorithms.