Search

This paper presents a sparse representation and an adaptive dictionary learning based method for automated classification of multiple sclerosis (MS) lesions in magnetic resonance (MR) images. Manual delineation of MS lesions is a time-consuming task, requiring neuroradiology experts to analyze huge volume of MR data. This, in addition to the high intra- and inter-observer variability necessitates the requirement of automated MS lesion classification methods. Among many image representation models and classification methods that can be used for such purpose, we investigate the use of sparse modeling. In the recent years, sparse representation has evolved as a tool in modeling data using a few basis elements of an over-complete dictionary and has found applications in many image processing tasks including classification. We propose a supervised classification approach by learning dictionaries specific to the lesions and individual healthy brain tissues, which include white matter (WM), gray matter (GM) and cerebrospinal fluid (CSF). The size of the dictionaries learned for each class plays a major role in data representation but it is an even more crucial element in the case of competitive classification. Our approach adapts the size of the dictionary for each class, depending on the complexity of the underlying data. The algorithm is validated using 52 multi-sequence MR images acquired from 13 MS patients. The results demonstrate the effectiveness of our approach in MS lesion classification. [ABSTRACT FROM AUTHOR]

Background: Non-Hispanic black (NHB) individuals have increased risk of Alzheimer's disease (AD) relative to non-Hispanic whites (NHW). Ethnicity/race can serve as a proxy sociodemographic variable for a complex representation of sociocultural and environmental factors. Chronic pain is a form of stress with high prevalence and sociodemographic disparities. Chronic pain is linked to lower cognition and accelerated biological aging.Objective: The purpose of this study is to seek understanding of potential cognitive and temporal lobe structural brain AD vulnerabilities based on chronic pain stage and ethnicity/race.Methods: Participants included 147 community dwelling NHB and NHW adults without dementia between 45-85 years old who had or were at risk of knee osteoarthritis. All participants received an MRI (3T Philips), the Montreal Cognitive Assessment (MoCA), and assessment of clinical knee pain stage.Results: There were ethnic/race group differences in MoCA scores but no relationships with chronic knee pain stage. Ethnicity/race moderated the relationship between AD-related temporal lobe thickness and chronic pain stage with quadratic patterns suggesting thinner cortex in high chronic pain stage NHB adults.Conclusion: There appear to be complex relationships between chronic knee pain stage, temporal lobe cortex, and sociodemographic variables. Specifically, NHB participants without dementia but with high chronic knee pain stage appeared to have thinner temporal cortex in areas associated with AD. Understanding the effects of sociocultural and socioeconomic factors on health outcomes is the first step to challenging the disparities in healthcare that now appear to link disease conditions to neurodegenerative processes. [ABSTRACT FROM AUTHOR]

Purpose: Deep brain stimulation is a common treatment for medication-refractory essential tremor. Current coordinate-based targeting methods result in variable outcomes due to variation in thalamic structure and the optimal patient-specific functional location. The purpose of this study was to compare the coordinate-based pre-operative targets to patient-specific thalamic segmentation utilizing a probabilistic tractography methodology.Methods: Using available diffusion MRI of 32 subjects from the Human Connectome Project database, probabilistic tractography was performed. Each thalamic voxel was coded based on one of six predefined cortical targets. The segmentation results were analyzed and compared to a 2-mm spherical target centered at the coordinate-based location of the ventral intermediate thalamic nucleus.Results: The traditional coordinate-based target had maximal overlap with the junction of the region most connected to primary motor cortex (M1) (36.6 ± 25.7% of voxels on left; 58.1 ± 28.5% on right) and the area connected to the supplementary motor area/premotor cortex (SMA/PMC) (44.9 ± 21.7% of voxels on left; 28.9 ± 22.2% on right). There was a within-subject coefficient of variation from right-to-left of 69.4 and 63.1% in the volume of overlap with the SMA/PMC and M1 regions, respectively.Conclusion: Thalamic segmentation based on structural connectivity measures is a promising technique that may enhance traditional targeting methods by generating reproducible, patient-specific pre-operative functional targets. Our results highlight the problematic intra- and inter-subject variability of indirect, coordinate-based targets. Future prospective clinical studies will be needed to validate this targeting methodology in essential tremor patients. [ABSTRACT FROM AUTHOR]

The social communication impairments defining autism spectrum disorders (ASD) may be built upon core deficits in perspective-taking, language processing, and self-other representation. Self-referential processing entails the ability to incorporate self-awareness, self-judgment, and self-memory in information processing. Very few studies have examined the neural bases of integrating self-other representation and semantic processing in individuals with ASD. The main objective of this functional MRI study is to examine the role of language and social brain networks in self-other processing in young adults with ASD. Nineteen high-functioning male adults with ASD and 19 age-sex-and-IQ-matched typically developing (TD) control participants made “yes” or “no” judgments of whether an adjective, presented visually, described them ( self ) or their favorite teacher ( other ). Both ASD and TD participants showed significantly increased activity in the medial prefrontal cortex (MPFC) during self and other processing relative to letter search. Analyses of group differences revealed significantly reduced activity in left inferior frontal gyrus (LIFG), and left inferior parietal lobule (LIPL) in ASD participants, relative to TD controls. ASD participants also showed significantly weaker functional connectivity of the anterior cingulate cortex (ACC) with several brain areas while processing self-related words. The LIFG and IPL are important regions functionally at the intersection of language and social roles; reduced recruitment of these regions in ASD participants may suggest poor level of semantic and social processing. In addition, poor connectivity of the ACC may suggest the difficulty in meeting the linguistic and social demands of this task in ASD. Overall, this study provides new evidence of the altered recruitment of the neural networks underlying language and social cognition in ASD. [ABSTRACT FROM AUTHOR]

In conjunction with the ISBI 2015 conference, we organized a longitudinal lesion segmentation challenge providing training and test data to registered participants. The training data consisted of five subjects with a mean of 4.4 time-points, and test data of fourteen subjects with a mean of 4.4 time-points. All 82 data sets had the white matter lesions associated with multiple sclerosis delineated by two human expert raters. Eleven teams submitted results using state-of-the-art lesion segmentation algorithms to the challenge, with ten teams presenting their results at the conference. We present a quantitative evaluation comparing the consistency of the two raters as well as exploring the performance of the eleven submitted results in addition to three other lesion segmentation algorithms. The challenge presented three unique opportunities: (1) the sharing of a rich data set; (2) collaboration and comparison of the various avenues of research being pursued in the community; and (3) a review and refinement of the evaluation metrics currently in use. We report on the performance of the challenge participants, as well as the construction and evaluation of a consensus delineation. The image data and manual delineations will continue to be available for download, through an evaluation website 2 2 The Challenge Evaluation Website is: http://smart-stats-tools.org/lesion-challenge-2015 as a resource for future researchers in the area. This data resource provides a platform to compare existing methods in a fair and consistent manner to each other and multiple manual raters. [ABSTRACT FROM AUTHOR]

While deficits in imitation had been reported in children with autism spectrum disorder (ASD), its exact nature remains unclear. A dysfunction in mirroring mechanisms (through action imitation) has been proposed by some studies to explain this, although some recent evidence points against this hypothesis. The current study used behavior and functional MRI to examine the integrated functioning of the regions that are considered part of the Action Imitation network (AIN) in children and adolescents with ASD during a motor imitation task. Fourteen ASD and 15 age-and-IQ-matched typically developing (TD) children were asked to imitate a series of hand gestures in the MRI scanner. Intact performance on imitation (accurate imitation of hand gestures outside the scanner) in both ASD and TD groups was accompanied by significantly lower activity in ASD participants, relative to TD, in right angular gyrus, precentral gyrus, and left middle cingulate. In addition, autism traits were found to be significantly correlated with activation in the right angular gyrus. Overall, the findings of this study support the role of AIN in imitation and a potential difference in the recruitment of this network in ASD children. [ABSTRACT FROM AUTHOR]