PMCC PMCC

Search tips
Search criteria

Advanced
Results 1-9 (9)
 

Clipboard (0)
None

Select a Filter Below

Journals
Year of Publication
Document Types
1.  Automated Diagnosis of Epilepsy using EEG Power Spectrum 
Epilepsia  2012;53(11):e189-e192.
Interictal electroencephalography (EEG) has clinically meaningful limitations in its sensitivity and specificity in the diagnosis of epilepsy because of its dependence on the occurrence of epileptiform discharges. We have developed a computer-aided diagnostic (CAD) tool that operates on the absolute spectral energy of the routine EEG and has both substantially higher sensitivity and negative predictive value than the identification of interictal epileptiform discharges. Our approach used a multilayer perceptron to classify 156 patients admitted for video-EEG monitoring. The patient population was diagnostically diverse with 87 diagnosed with either generalized or focal seizures. The remainder was diagnosed with non-epileptic seizures. The sensitivity was 92% (95% CI: 85–97%) and the negative predictive value was 82% (95% CI: 67%–92%). We discuss how these findings suggest that this CAD can be used to supplement event-based analysis by trained epileptologists.
doi:10.1111/j.1528-1167.2012.03653.x
PMCID: PMC3447367  PMID: 22967005
Epilepsy; machine learning; prediction; non-epileptic seizure; computer aided diagnostics
2.  Computer-Aided Diagnosis and Localization of Lateralized Temporal Lobe Epilepsy Using Interictal FDG-PET 
Interictal FDG-PET (iPET) is a core tool for localizing the epileptogenic focus, potentially before structural MRI, that does not require rare and transient epileptiform discharges or seizures on EEG. The visual interpretation of iPET is challenging and requires years of epilepsy-specific expertise. We have developed an automated computer-aided diagnostic (CAD) tool that has the potential to work both independent of and synergistically with expert analysis. Our tool operates on distributed metabolic changes across the whole brain measured by iPET to both diagnose and lateralize temporal lobe epilepsy (TLE). When diagnosing left TLE (LTLE) or right TLE (RTLE) vs. non-epileptic seizures (NES), our accuracy in reproducing the results of the gold standard long term video-EEG monitoring was 82% [95% confidence interval (CI) 69–90%] or 88% (95% CI 76–94%), respectively. The classifier that both diagnosed and lateralized the disease had overall accuracy of 76% (95% CI 66–84%), where 89% (95% CI 77–96%) of patients correctly identified with epilepsy were correctly lateralized. When identifying LTLE, our CAD tool utilized metabolic changes across the entire brain. By contrast, only temporal regions and the right frontal lobe cortex, were needed to identify RTLE accurately, a finding consistent with clinical observations and indicative of a potential pathophysiological difference between RTLE and LTLE. The goal of CADs is to complement – not replace – expert analysis. In our dataset, the accuracy of manual analysis (MA) of iPET (∼80%) was similar to CAD. The square correlation between our CAD tool and MA, however, was only 30%, indicating that our CAD tool does not recreate MA. The addition of clinical information to our CAD, however, did not substantively change performance. These results suggest that automated analysis might provide clinically valuable information to focus treatment more effectively.
doi:10.3389/fneur.2013.00031
PMCID: PMC3615243  PMID: 23565107
epilepsy; computer-aided diagnosis; mutual information; temporal lobe epilepsy; PET; fluoro-deoxyglucose positron emission tomography; machine learning
3.  Single trial decoding of belief decision making from EEG and fMRI data using independent components features 
The complex task of assessing the veracity of a statement is thought to activate uniquely distributed brain regions based on whether a subject believes or disbelieves a given assertion. In the current work, we present parallel machine learning methods for predicting a subject's decision response to a given propositional statement based on independent component (IC) features derived from EEG and fMRI data. Our results demonstrate that IC features outperformed features derived from event related spectral perturbations derived from any single spectral band, yet were similar to accuracy across all spectral bands combined. We compared our diagnostic IC spatial maps with our conventional general linear model (GLM) results, and found that informative ICs had significant spatial overlap with our GLM results, yet also revealed unique regions like amygdala that were not statistically significant in GLM analyses. Overall, these results suggest that ICs may yield a parsimonious feature set that can be used along with a decision tree structure for interpretation of features used in classifying complex cognitive processes such as belief and disbelief across both fMRI and EEG neuroimaging modalities.
doi:10.3389/fnhum.2013.00392
PMCID: PMC3728485  PMID: 23914164
machine learning; decoding; EEG; fMRI; ICA; decision making; decision tree; interpretation
4.  fMRI of syntactic processing in typically developing children: structural correlates in the inferior frontal gyrus 
Development of syntactic processing was examined to evaluate maturational processes including left language lateralization functions and increased specialization of brain regions important for syntactic processing. We utilized multimodal methods, including indices of brain activity from fMRI during a syntactic processing task, cortical thickness measurements from structural MRI, and neuropsychological measures. To evaluate hypotheses about increasing lateralization and specialization with development, we examined relationships between cortical thickness and magnitude and spatial activation extent within the left inferior frontal gyrus (IFG) and its right hemisphere homologue. We predicted that increased activation in the left and decreased activation in the right IFG would be associated with increased syntactic proficiency. As predicted, a more mature pattern of increased thickness in the right pars triangularis was associated with decreased activation intensity and extent in the right IFG. These findings suggest a maturational shift towards decreased involvement of the right IFG for syntactic processing. Better syntactic skills were associated with increased activation in the left IFG independent from age, suggesting increased specialization of the left IFG with increased proficiency. Overall, our findings show relationships between structural and functional neurodevelopment that co-occur with improved syntactic processing in critical language regions of the IFG in typically developing children.
doi:10.1016/j.dcn.2011.02.004
PMCID: PMC3129989  PMID: 21743820
Syntax; language; typical development; lateralization; fMRI; multimodal
5.  Sex Matters during Adolescence: Testosterone-Related Cortical Thickness Maturation Differs between Boys and Girls 
PLoS ONE  2012;7(3):e33850.
Age-related changes in cortical thickness have been observed during adolescence, including thinning in frontal and parietal cortices, and thickening in the lateral temporal lobes. Studies have shown sex differences in hormone-related brain maturation when boys and girls are age-matched, however, because girls mature 1–2 years earlier than boys, these sex differences could be confounded by pubertal maturation. To address puberty effects directly, this study assessed sex differences in testosterone-related cortical maturation by studying 85 boys and girls in a narrow age range and matched on sexual maturity. We expected that testosterone-by-sex interactions on cortical thickness would be observed in brain regions known from the animal literature to be high in androgen receptors. We found sex differences in associations between circulating testosterone and thickness in left inferior parietal lobule, middle temporal gyrus, calcarine sulcus, and right lingual gyrus, all regions known to be high in androgen receptors. Visual areas increased with testosterone in boys, but decreased in girls. All other regions were more impacted by testosterone levels in girls than boys. The regional pattern of sex-by-testosterone interactions may have implications for understanding sex differences in behavior and adolescent-onset neuropsychiatric disorders.
doi:10.1371/journal.pone.0033850
PMCID: PMC3315517  PMID: 22479458
6.  Puberty Influences Medial Temporal Lobe and Cortical Gray Matter Maturation Differently in Boys Than Girls Matched for Sexual Maturity 
Cerebral Cortex (New York, NY)  2010;21(3):636-646.
Sex differences in age- and puberty-related maturation of human brain structure have been observed in typically developing age-matched boys and girls. Because girls mature 1–2 years earlier than boys, the present study aimed at assessing sex differences in brain structure by studying 80 adolescent boys and girls matched on sexual maturity, rather than age. We evaluated pubertal influences on medial temporal lobe (MTL), thalamic, caudate, and cortical gray matter volumes utilizing structural magnetic resonance imaging and 2 measures of pubertal status: physical sexual maturity and circulating testosterone. As predicted, significant interactions between sex and the effect of puberty were observed in regions with high sex steroid hormone receptor densities; sex differences in the right hippocampus, bilateral amygdala, and cortical gray matter were greater in more sexually mature adolescents. Within sex, we found larger volumes in MTL structures in more sexually mature boys, whereas smaller volumes were observed in more sexually mature girls. Our results demonstrate puberty-related maturation of the hippocampus, amygdala, and cortical gray matter that is not confounded by age, and is different for girls and boys, which may contribute to differences in social and cognitive development during adolescence, and lasting sexual dimorphisms in the adult brain.
doi:10.1093/cercor/bhq137
PMCID: PMC3041011  PMID: 20713504
7.  Effect of Bupropion Treatment on Brain Activation Induced by Cigarette-Related Cues in Smokers 
Archives of general psychiatry  2011;68(5):505-515.
Context
Nicotine-dependent smokers exhibit craving and brain activation in the prefrontal and limbic regions when presented with cigarette-related cues. Bupropion hydrochloride treatment reduces cue-induced craving in cigarette smokers; however, the mechanism by which bupropion exerts this effect has not yet been described.
Objective
To assess changes in regional brain activation in response to cigarette-related cues from before to after treatment with bupropion (vs placebo).
Design
Randomized, double-blind, before-after controlled trial.
Setting
Academic brain imaging center.
Participants
Thirty nicotine-dependent smokers (paid volunteers).
Interventions
Participants were randomly assigned to receive 8 weeks of treatment with either bupropion or a matching placebo pill (double-blind).
Main Outcome Measures
Subjective cigarette craving ratings and regional brain activations (blood oxygen level-dependent response) in response to viewing cue videos.
Results
Bupropion-treated participants reported less craving and exhibited reduced activation in the left ventral striatum, right medial orbitofrontal cortex, and bilateral anterior cingulate cortex from before to after treatment when actively resisting craving compared with placebo-treated participants. When resisting craving, reduction in self-reported craving correlated with reduced regional brain activation in the bilateral medial orbitofrontal and left anterior cingulate cortices in all participants.
Conclusions
Treatment with bupropion is associated with improved ability to resist cue-induced craving and a reduction in cue-induced activation of limbic and prefrontal brain regions, while a reduction in craving, regardless of treatment type, is associated with reduced activation in prefrontal brain regions.
doi:10.1001/archgenpsychiatry.2010.193
PMCID: PMC3214639  PMID: 21199957
8.  Large Sample Group Independent Component Analysis of Functional Magnetic Resonance Imaging Using Anatomical Atlas-Based Reduction and Bootstrapped Clustering 
Independent component analysis (ICA) is a popular method for the analysis of functional magnetic resonance imaging (fMRI) signals that is capable of revealing connected brain systems of functional significance. To be computationally tractable, estimating the independent components (ICs) inevitably requires one or more dimension reduction steps. Whereas most algorithms perform such reductions in the time domain, the input data are much more extensive in the spatial domain, and there is broad consensus that the brain obeys rules of localization of function into regions that are smaller in number than the number of voxels in a brain image. These functional units apparently reorganize dynamically into networks under different task conditions. Here we develop a new approach to ICA, producing group results by bagging and clustering over hundreds of pooled single-subject ICA results that have been projected to a lower-dimensional subspace. Averages of anatomically based regions are used to compress the single subject-ICA results prior to clustering and resampling via bagging. The computational advantages of this approach make it possible to perform group-level analyses on datasets consisting of hundreds of scan sessions by combining the results of within-subject analysis, while retaining the theoretical advantage of mimicking what is known of the functional organization of the brain. The result is a compact set of spatial activity patterns that are common and stable across scan sessions and across individuals. Such representations may be used in the context of statistical pattern recognition supporting real-time state classification.
doi:10.1002/ima.20286
PMCID: PMC3204794  PMID: 22049263
fMRI; group ICA; bagging; clustering; bootstrap
9.  Influence of the interstimulus interval on temporal processing and learning: testing the state-dependent network model 
The ability to determine the interval and duration of sensory events is fundamental to most forms of sensory processing, including speech and music perception. Recent experimental data support the notion that different mechanisms underlie temporal processing in the subsecond and suprasecond range. Here, we examine the predictions of one class of subsecond timing models: state-dependent networks. We establish that the interval between the comparison and the test interval, interstimulus interval (ISI), in a two-interval forced-choice discrimination task, alters the accuracy of interval discrimination but not the point of subjective equality—i.e. while timing was impaired, subjective time contraction or expansion was not observed. We also examined whether the deficit in temporal processing produced by short ISIs can be reduced by learning, and determined the generalization patterns. These results show that training subjects on a task using a short or long ISI produces dramatically different generalization patterns, suggesting different forms of perceptual learning are being engaged. Together, our results are consistent with the notion that timing in the range of hundreds of milliseconds is local as opposed to centralized, and that rapid stimulus presentation rates impair temporal discrimination. This interference is, however, decreased if the stimuli are presented to different sensory channels.
doi:10.1098/rstb.2009.0019
PMCID: PMC2685819  PMID: 19487189
timing; interval discrimination; perceptual learning; temporal processing; generalization

Results 1-9 (9)