Consistent segmentation using a Rician classifier

Several popular classification algorithms used to segment magnetic resonance brain images assume that the image intensities, or log-transformed intensities, satisfy a finite Gaussian mixture model. In these methods, the parameters of the mixture model are estimated and the posterior probabilities for each tissue class are used directly as soft segmentations or combined to form a hard segmentation. It is suggested and shown in this paper that a Rician mixture model fits the observed data better than a Gaussian model. Accordingly, a Rician mixture model is formulated and used within an expectation maximization (EM) framework to yield a new tissue classification algorithm called Rician Classifier using EM (RiCE). It is shown using both simulated and real data that RiCE yields comparable or better performance to that of algorithms based on the finite Gaussian mixture model. As well, we show that RiCE yields more consistent segmentation results when used on images of the same individual acquired with different T1-weighted pulse sequences. Therefore, RiCE has the potential to stabilize segmentation results in brain studies involving heterogeneous acquisition sources as is typically found in both multi-center and longitudinal studies.

A compressed sensing approach for MR tissue contrast synthesis

The tissue contrast of a magnetic resonance (MR) neuroimaging data set has a major impact on image analysis tasks like registration and segmentation. It has been one of the core challenges of medical imaging to guarantee the consistency of these tasks regardless of the contrasts of the MR data. Inconsistencies in image analysis are attributable in part to variations in tissue contrast, which in turn arise from operator variations during image acquisition as well as software and hardware differences in the MR scanners. It is also a common problem that images with a desired tissue contrast are completely missing in a given data set for reasons of cost, acquisition time, forgetfulness, or patient comfort. Absence of this data can hamper the detailed, automatic analysis of some or all data sets in a scientific study. A method to synthesize missing MR tissue contrasts from available acquired images using an atlas containing the desired contrast and a patch-based compressed sensing strategy is described. An important application of this general approach is to synthesize a particular tissue contrast from multiple studies using a single atlas, thereby normalizing all data sets into a common intensity space. Experiments on real data, obtained using different scanners and pulse sequences, show improvement in segmentation consistency, which could be extremely valuable in the pooling of multi-site multi-scanner neuroimaging studies.

Simple paradigm for extra-cerebral tissue removal: algorithm and analysis

Extraction of the brain-i.e. cerebrum, cerebellum, and brain stem-from T1-weighted structural magnetic resonance images is an important initial step in neuroimage analysis. Although automatic algorithms are available, their inconsistent handling of the cortical mantle often requires manual interaction, thereby reducing their effectiveness. This paper presents a fully automated brain extraction algorithm that incorporates elastic registration, tissue segmentation, and morphological techniques which are combined by a watershed principle, while paying special attention to the preservation of the boundary between the gray matter and the cerebrospinal fluid. The approach was evaluated by comparison to a manual rater, and compared to several other leading algorithms on a publically available data set of brain images using the Dice coefficient and containment index as performance metrics. The qualitative and quantitative impact of this initial step on subsequent cortical surface generation is also presented. Our experiments demonstrate that our approach is quantitatively better than six other leading algorithms (with statistical significance on modern T1-weighted MR data). We also validated the robustness of the algorithm on a very large data set of over one thousand subjects, and showed that it can replace an experienced manual rater as preprocessing for a cortical surface extraction algorithm with statistically insignificant differences in cortical surface position.

A framework on surface-based connectivity quantification for the human brain

Quantifying the connectivity between arbitrary surface patches in the human brain cortex can be used in studies on brain function and to characterize clinical diseases involving abnormal connectivity. Cortical regions of human brain in their natural forms can be represented in surface formats. In this paper, we present a framework to quantify connectivity using cortical surface segmentation and labeling from structural magnetic resonance images, tractography from diffusion tensor images, and nonlinear inter-subject registration. For a single subject, the connectivity intensity of any point on the cortical surface is set to unity if the point is connected and zero if it is not connected. The connectivity proportion is defined as the ratio of the total connected surface area to the total area of the surface patch. By nonlinearly registering the connectivity data of a group of normal controls into a template space, a population connectivity metric can be defined as either the average connectivity intensity of a cortical point or the average connectivity proportion of a cortical region. In the template space, a connectivity profile and a connectivity histogram of an arbitrary cortical region of interest can then be derived from these connectivity quantification values. Results from the application of these quantification metrics to a population of schizophrenia patients and normal controls are presented, revealing connectivity signatures of specified cortical regions and detecting connectivity abnormalities.

Distance Transforms in Multi Channel MR Image Registration

Deformable registration techniques play vital roles in a variety of medical imaging tasks such as image fusion, segmentation, and post-operative surgery assessment. In recent years, mutual information has become one of the most widely used similarity metrics for medical image registration algorithms. Unfortunately, as a matching criteria, mutual information loses much of its effectiveness when there is poor statistical consistency and a lack of structure. This is especially true in areas of images where the intensity is homogeneous and information is sparse. Here we present a method designed to address this problem by integrating distance transforms of anatomical segmentations as part of a multi-channel mutual information framework within the registration algorithm. Our method was tested by registering real MR brain data and comparing the segmentation of the results against that of the target. Our analysis showed that by integrating distance transforms of the the white matter segmentation into the registration, the overall segmentation of the registration result was closer to the target than when the distance transform was not used.

Intensity Inhomogeneity Correction of Magnetic Resonance Images using Patches

This paper presents a patch-based non-parametric approach to the correction of intensity inhomogeneity from magnetic resonance (MR) images of the human brain. During image acquisition, the inhomogeneity present in the radio-frequency coil, is usually manifested on the reconstructed MR image as a smooth shading effect. This artifact can significantly deteriorate the performance of any kind of image processing algorithm that uses intensities as a feature. Most of the current inhomogeneity correction techniques use explicit smoothness assumptions on the inhomogeneity field, which sometimes limit their performance if the actual inhomogeneity is not smooth, a problem that becomes prevalent in high fields. The proposed patch-based inhomogeneity correction method does not assume any parametric smoothness model, instead, it uses patches from an atlas of an inhomogeneity-free image to do the correction. Preliminary results show that the proposed method is comparable to N3, a current state of the art method, when the inhomogeneity is smooth, and outperforms N3 when the inhomogeneity contains non-smooth elements.

TOPOLOGY PRESERVING AUTOMATIC SEGMENTATION OF THE SPINAL CORD IN MAGNETIC RESONANCE IMAGES

Magnetic resonance images of the spinal cord play an important role in studying neurological diseases, particularly multiple sclerosis, where spinal cord atrophy can provide a measure of disease progression and disability. Current practices involve segmenting the spinal cord manually, which can be an inconsistent and time-consuming process. We present an automatic segmentation method for the spinal cord using a combination of deformable atlas based registration and topology preserving classification. Using real MR data, our method is shown to be highly accurate when compared to segmentations by manual raters. In addition, our results always maintain the correct topology of the spinal cord, therefore providing segmentations more consistent with the known anatomy.

COMPRESSED SENSING BASED INTENSITY NON-UNIFORMITY CORRECTION

We present a compressed sensing based approach to remove gain field from magnetic resonance (MR) images of the human brain. During image acquisition, the inhomogeneity present in the radio-frequency (RF) coil appears as shading artifact in the intensity image. The inhomogeneity poses problem in any automatic algorithm that uses intensity as a feature. It has been shown that at low field strength, the shading can be assumed to be a smooth field that is composed of low frequency components. Thus most inhomogeneity correction algorithms assume some kind of explicit smoothness criteria on the field. This sometimes limits the performance of the algorithms if the actual inhomogeneity is not smooth, which is the case at higher field strength. We describe a model-free, non-parametric patch-based approach that uses compressed sensing for the correction. We show that these features enable our algorithm to perform comparably with a current state of the art method N3 on images acquired at low field, while outperforming N3 when the image has non-smooth inhomogeneity, such as 7T images.

The Java Image Science Toolkit (JIST) for rapid prototyping and publishing of neuroimaging software

Non-invasive neuroimaging techniques enable extraordinarily sensitive and specific in vivo study of the structure, functional response and connectivity of biological mechanisms. With these advanced methods comes a heavy reliance on computer-based processing, analysis and interpretation. While the neuroimaging community has produced many excellent academic and commercial tool packages, new tools are often required to interpret new modalities and paradigms. Developing custom tools and ensuring interoperability with existing tools is a significant hurdle. To address these limitations, we present a new framework for algorithm development that implicitly ensures tool interoperability, generates graphical user interfaces, provides advanced batch processing tools, and, most importantly, requires minimal additional programming or computational overhead. Java-based rapid prototyping with this system is an efficient and practical approach to evaluate new algorithms since the proposed system ensures that rapidly constructed prototypes are actually fully-functional processing modules with support for multiple GUI's, a broad range of file formats, and distributed computation. Herein, we demonstrate MRI image processing with the proposed system for cortical surface extraction in large cross-sectional cohorts, provide a system for fully automated diffusion tensor image analysis, and illustrate how the system can be used as a simulation framework for the development of a new image analysis method. The system is released as open source under the Lesser GNU Public License (LGPL) through the Neuroimaging Informatics Tools and Resources Clearinghouse (NITRC).

Longitudinal changes in cortical thickness associated with normal aging

Imaging studies of anatomic changes in regional gray matter volumes and cortical thickness have documented age effects in many brain regions, but the majority of such studies have been cross-sectional investigations of individuals studied at a single point in time. In this study, using serial imaging assessments of participants in the Baltimore Longitudinal Study of Aging (BLSA), we investigate longitudinal changes in cortical thickness during aging in a cohort of 66 older adults (mean age 68.78; sd. 6.6; range 60-84 at baseline) without dementia. We used the Cortical Reconstruction Using Implicit Surface Evolution CRUISE suite of algorithms to automatically generate a reconstruction of the cortical surface and identified twenty gyral based regions of interest per hemisphere. Using mixed effects regression, we investigated longitudinal changes in these regions over a mean follow-up interval of 8 years. The main finding in this study is that age-related decline in cortical thickness is widespread, but shows an anterior-posterior gradient with frontal and parietal regions, in general, exhibiting greater rates of decline than temporal and occipital. There were fewer regions in the right hemisphere showing statistically significant age-associated longitudinal decreases in mean cortical thickness. Males showed greater rates of decline in the middle frontal, inferior parietal, parahippocampal, postcentral, and superior temporal gyri in the left hemisphere, right precuneus and bilaterally in the superior parietal and cingulate regions. Significant nonlinear changes over time were observed in the postcentral, precentral, and orbitofrontal gyri on the left and inferior parietal, cingulate, and orbitofrontal gyri on the right.

Interfaces and Integration of Medical Image Analysis Frameworks: Challenges and Opportunities

Clinical research with medical imaging typically involves large-scale data analysis with interdependent software toolsets tied together in a processing workflow. Numerous, complementary platforms are available, but these are not readily compatible in terms of workflows or data formats. Both image scientists and clinical investigators could benefit from using the framework which is a most natural fit to the specific problem at hand, but pragmatic choices often dictate that a compromise platform is used for collaboration. Manual merging of platforms through carefully tuned scripts has been effective, but exceptionally time consuming and is not feasible for large-scale integration efforts. Hence, the benefits of innovation are constrained by platform dependence. Removing this constraint via integration of algorithms from one framework into another is the focus of this work. We propose and demonstrate a light-weight interface system to expose parameters across platforms and provide seamless integration. In this initial effort, we focus on four platforms Medical Image Analysis and Visualization (MIPAV), Java Image Science Toolkit (JIST), command line tools, and 3D Slicer. We explore three case studies: (1) providing a system for MIPAV to expose internal algorithms and utilize these algorithms within JIST, (2) exposing JIST modules through self-documenting command line interface for inclusion in scripting environments, and (3) detecting and using JIST modules in 3D Slicer. We review the challenges and opportunities for light-weight software integration both within development language (e.g., Java in MIPAV and JIST) and across languages (e.g., C/C++ in 3D Slicer and shell in command line tools).

Synthesizing MR Contrast and Resolution through a Patch Matching Technique

Tissue contrast and resolution of magnetic resonance neuroimaging data have strong impacts on the utility of the data in clinical and neuroscience tasks such as registration and segmentation. Lengthy acquisition times typically prevent routine acquisition of multiple MR tissue contrast images at high resolution, and the opportunity for detailed analysis using these data would seem to be irrevocably lost. This paper describes an example based approach using patch matching from a multiple resolution multiple contrast atlas in order to change an image's resolution as well as its MR tissue contrast from one pulse-sequence to that of another. The use of this approach to generate different tissue contrasts (T2/PD/FLAIR) from a single T1-weighted image is demonstrated on both phantom and real images.

MR CONTRAST SYNTHESIS FOR LESION SEGMENTATION

The magnetic resonance contrast of a neuroimaging data set has strong impact on the utility of the data in image analysis tasks, such as registration and segmentation. Lengthy acquisition times often prevent routine acquisition of multiple MR contrast images, and opportunities for detailed analysis using these data would seem to be irrevocably lost. This paper describes an example based approach which uses patch matching from a multiple contrast atlas with the intended goal of generating an alternate MR contrast image, thus effectively simulating alternative pulse sequences from one another. In this paper, we deal specifically with Fluid Attenuated Inversion Recovery (FLAIR) sequence generation from T1 and T2 pulse sequences. The applicability of this synthetic FLAIR for estimating white matter lesions segmentation is demonstrated.

A RICIAN MIXTURE MODEL CLASSIFICATION ALGORITHM FOR MAGNETIC RESONANCE IMAGES

Tissue classification algorithms developed for magnetic resonance images commonly assume a Gaussian model on the statistics of noise in the image. While this is approximately true for voxels having large intensities, it is less true as the underlying intensity becomes smaller. In this paper, the Gaussian model is replaced with a Rician model, which is a better approximation to the observed signal. A new classification algorithm based on a finite mixture model of Rician signals is presented wherein the expectation maximization algorithm is used to find the joint maximum likelihood estimates of the unknown mixture parameters. Improved accuracy of tissue classification is demonstrated on several sample data sets. It is also shown that classification repeatability for the same subject under different MR acquisitions is improved using the new method.

AUTOMATED RELIABLE LABELING OF THE CORTICAL SURFACE

Accurate and reliable labeling of regions of interest (ROIs) within structural magnetic resonance images (MRIs) is an important step in group analysis. Group registration does not always yield accurate alignment of homologous regions. We present an approach that distinguishes itself from other algorithms by being concerned with a label which is of the highest fidelity, while leaving ambiguous regions unlabeled. Regions that are not deemed to be reliably labeled are not included in group statistics. We will present results showing that our method is an improvement over traditional multi-atlas voting schemes. We conclude with a pilot study of longitudinal trends of cortical thickness in normal aging.

AUTOMATICALLY IDENTIFYING WHITE MATTER TRACTS USING CORTICAL LABELS

Diffusion tensor imaging (DTI) has become a standard clinical procedure in assessing the health of white matter in the brain. Tractography, the tracing of individual fibers in the brain using DTI data, has begun to play a more central role in neuroscience research, particularly in understanding the relationships between brain connectivity and behavior. The measuring of features related to bundles of fibers, i.e., tracts or fasciculi, is currently problematic because of the need for manual interaction. This article presents an algorithm for the automatic identification of selected white matter tracts. It extracts fibers using the FACT algorithm and finds cortical gyral labels using a multi-atlas deformable registration scheme. Tracts are identified as the fibers passing between selected cortical labels. The quality of automatic labels are compared both visually and numerically against a well-accepted manual approach. The automatic approach is shown to be more consistent with conventional definitions of tracts and more repeatable on separate scans of the same subject.

FUZZY C-MEANS WITH VARIABLE COMPACTNESS

Fuzzy c-means (FCM) clustering has been extensively studied and widely applied in the tissue classification of biomedical images. Previous enhancements to FCM have accounted for intensity shading, membership smoothness, and variable cluster sizes. In this paper, we introduce a new parameter called "compactness" which captures additional information of the underlying clusters. We then propose a new classification algorithm, FCM with variable compactness (FCMVC), to classify three major tissues in brain MRIs by incorporating the compactness terms into a previously reported improvement to FCM. Experiments on both simulated phantoms and real magnetic resonance brain images show that the new method improves the repeatability of the tissue classification for the same subject with different acquisition protocols.