Automated Anatomical Labeling Research Articles

A Deep Learning Approach to Predict Autism Spectrum Disorder Using Multisite Resting-State fMRI

Autism spectrum disorder (ASD) is a complex and degenerative neuro-developmental disorder. Most of the existing methods utilize functional magnetic resonance imaging (fMRI) to detect ASD with a very limited dataset which provides high accuracy but results in poor generalization. To overcome this limitation and to enhance the performance of the automated autism diagnosis model, in this paper, we propose an ASD detection model using functional connectivity features of resting-state fMRI data. Our proposed model utilizes two commonly used brain atlases, Craddock 200 (CC200) and Automated Anatomical Labelling (AAL), and two rarely used atlases Bootstrap Analysis of Stable Clusters (BASC) and Power. A deep neural network (DNN) classifier is used to perform the classification task. Simulation results indicate that the proposed model outperforms state-of-the-art methods in terms of accuracy. The mean accuracy of the proposed model was 88%, whereas the mean accuracy of the state-of-the-art methods ranged from 67% to 85%. The sensitivity, F1-score, and area under receiver operating characteristic curve (AUC) score of the proposed model were 90%, 87%, and 96%, respectively. Comparative analysis on various scoring strategies show the superiority of BASC atlas over other aforementioned atlases in classifying ASD and control.

Glucose metabolism in the right middle temporal gyrus could be a potential biomarker for subjective cognitive decline: a study of a Han population

IntroductionSubjective cognitive decline (SCD) represents a cognitively normal state but at an increased risk for developing Alzheimer’s disease (AD). Recognizing the glucose metabolic biomarkers of SCD could facilitate the location of areas with metabolic changes at an ultra-early stage. The objective of this study was to explore glucose metabolic biomarkers of SCD at the region of interest (ROI) level.MethodsThis study was based on cohorts from two tertiary medical centers, and it was part of the SILCODE project (NCT03370744). Twenty-six normal control (NC) cases and 32 SCD cases were in cohort 1; 36 NCs, 23 cases of SCD, 32 cases of amnestic mild cognitive impairment (aMCIs), 32 cases of AD dementia (ADDs), and 22 cases of dementia with Lewy bodies (DLBs) were in cohort 2. Each subject underwent [18F]fluoro-2-deoxyglucose positron emission tomography (PET) imaging and magnetic resonance imaging (MRI), and subjects from cohort 1 additionally underwent amyloid-PET scanning. The ROI analysis was based on the Anatomical Automatic Labeling (AAL) template; multiple permutation tests and repeated cross-validations were conducted to determine the metabolic differences between NC and SCD cases. In addition, receiver operating characteristic curves were used to evaluate the capabilities of potential glucose metabolic biomarkers in distinguishing different groups. Pearson correlation analysis was also performed to explore the correlation between glucose metabolic biomarkers and neuropsychological scales or amyloid deposition.ResultsOnly the right middle temporal gyrus (RMTG) passed the methodological verification, and its metabolic levels were correlated with the degrees of complaints (R = − 0.239, p = 0.009), depression (R = − 0.200, p = 0.030), and abilities of delayed memory (R = 0.207, p = 0.025), and were weakly correlated with cortical amyloid deposition (R = − 0.246, p = 0.066). Furthermore, RMTG metabolism gradually decreased across the cognitive continuum, and its diagnostic efficiency was comparable (NC vs. ADD, aMCI, or DLB) or even superior (NC vs. SCD) to that of the metabolism of the posterior cingulate cortex or precuneus.ConclusionsThese findings suggest that the hypometabolism of RMTG could be a typical feature of SCD, and the large-scale hypometabolism in patients with symptomatic stages of AD may start from the RMTG, which gradually progresses starting in the preclinical stage. The specificity of identifying SCD from the perspective of self-perceived symptoms is likely to be increased by the detection of RMTG metabolism.

Interictal 18F-FDG brain PET metabolism in patients with postictal EEG suppression

ObjectivePostictal generalized suppression (PGES) may be associated with SUDEP risk. We aimed to study metabolic changes on 18Fluorodeoxyglucose (18F-FDG) positron emission tomography (PET) in patients with focal to bilateral (generalized) seizures (GTCS) and PGES on stereoelectroencephalography (SEEG). MethodsWe analyzed interictal brain metabolism in a group of 19 patients with widespread postictal suppression (PGES+) associated with SEEG-recorded GTCS. This group was compared to 25 patients without widespread suppression (PGES−) as defined by SEEG, matched for epilepsy localization and lateralization. Frequency of GTCS was observed to be higher in the PGES+ group (high risk group for SUDEP). Analysis of metabolic data was performed by statistical parametric mapping (SPM) on the whole-brain, and principal component analysis (PCA) on AAL (automated anatomical labeling) atlas. ResultsStatistical parametric mapping showed right temporal pole hypometabolism in the PGES+ group (T-score = 3.90; p < 0.001; k = 185), in comparison to the PGES− group. Principal component analysis showed association between the metabolic values of certain regions of interest and PGES+/PGES− groups, confirmed by a significant difference (p < 0.05) in the values of the right dorsal temporal pole and of the left temporal pole between the two groups. Principal component analysis showed two dimensions significantly related to the PGES+/PGES− partition, involving the following regions: right temporal pole, right parahippocampal gyrus, right Rolandic operculum, bilateral paracentral lobule, right precuneus, right thalamus, right caudate and pallidum, bilateral cerebellum, left temporal pole, left Heschl’s gyrus, left calcarine region, and left caudate, with loss of connection in PGES+ patients. Metabolic differences were independent of epilepsy localization and lateralization and persisted after correction for GTCS frequency. SignificanceInterictal metabolic changes within a predominantly right-sided network involving temporal lobe and connected cortical and subcortical structures were seen in patients with frequent GTCS presenting widespread postictal suppression.

Repeatability and Reproducibility of in-vivo Brain Temperature Measurements.

Background: Magnetic resonance spectroscopic imaging (MRSI) is a neuroimaging technique that may be useful for non-invasive mapping of brain temperature (i.e., thermometry) over a large brain volume. To date, intra-subject reproducibility of MRSI-based brain temperature (MRSI-t) has not been investigated. The objective of this repeated measures MRSI-t study was to establish intra-subject reproducibility and repeatability of brain temperature, as well as typical brain temperature range.Methods: Healthy participants aged 23–46 years (N = 18; 7 females) were scanned at two time points ~12-weeks apart. Volumetric MRSI data were processed by reconstructing metabolite and water images using parametric spectral analysis. Brain temperature was derived using the frequency difference between water and creatine (TCRE) for 47 regions of interest (ROIs) delineated by the modified Automated Anatomical Labeling (AAL) atlas. Reproducibility was measured using the coefficient of variation for repeated measures (COVrep), and repeatability was determined using the standard error of measurement (SEM). For each region, the upper and lower bounds of Minimal Detectable Change (MDC) were established to characterize the typical range of TCRE values.Results: The mean global brain temperature over all subjects was 37.2°C with spatial variations across ROIs. There was a significant main effect for time [F(1, 1,591) = 37.0, p < 0.0001] and for brain region [F(46, 1,591) = 2.66, p < 0.0001]. The time*brain region interaction was not significant [F(46, 1,591) = 0.80, p = 0.83]. Participants' TCRE was stable for each ROI across both time points, with ROIs' COVrep ranging from 0.81 to 3.08% (mean COVrep = 1.92%); majority of ROIs had a COVrep <2.0%.Conclusions: Brain temperature measurements were highly consistent between both time points, indicating high reproducibility and repeatability of MRSI-t. MRSI-t may be a promising diagnostic, prognostic, and therapeutic tool for non-invasively monitoring brain temperature changes in health and disease. However, further studies of healthy participants with larger sample size(s) and numerous repeated acquisitions are imperative for establishing a reference range of typical brain TCRE, as well as the threshold above which TCRE is likely pathological.

Brain parcellation driven by dynamic functional connectivity better capture intrinsic network dynamics.

Until now, dynamic functional connectivity (dFC) based on functional magnetic resonance imaging is typically estimated on a set of predefined regions of interest (ROIs) derived from an anatomical or static functional atlas which follows an implicit assumption of functional homogeneity within ROIs underlying temporal fluctuation of functional coupling, potentially leading to biases or underestimation of brain network dynamics. Here, we presented a novel computational method based on dynamic functional connectivity degree (dFCD) to derive meaningful brain parcellations that can capture functional homogeneous regions in temporal variance of functional connectivity. Several spatially distributed but functionally meaningful areas that are well consistent with known intrinsic connectivity networks were identified through independent component analysis (ICA) of time‐varying dFCD maps. Furthermore, a systematical comparison with commonly used brain atlases, including the Anatomical Automatic Labeling template, static ICA‐driven parcellation and random parcellation, demonstrated that the ROI‐definition strategy based on the proposed dFC‐driven parcellation could better capture the interindividual variability in dFC and predict observed individual cognitive performance (e.g., fluid intelligence, cognitive flexibility, and sustained attention) based on chronnectome. Together, our findings shed new light on the functional organization of resting brains at the timescale of seconds and emphasized the significance of a dFC‐driven and voxel‐wise functional homogeneous parcellation for network dynamics analyses in neuroscience.

Identification of early mild cognitive impairment using multi-modal data and graph convolutional networks

BackgroundThe identification of early mild cognitive impairment (EMCI), which is an early stage of Alzheimer’s disease (AD) and is associated with brain structural and functional changes, is still a challenging task. Recent studies show great promises for improving the performance of EMCI identification by combining multiple structural and functional features, such as grey matter volume and shortest path length. However, extracting which features and how to combine multiple features to improve the performance of EMCI identification have always been a challenging problem. To address this problem, in this study we propose a new EMCI identification framework using multi-modal data and graph convolutional networks (GCNs). Firstly, we extract grey matter volume and shortest path length of each brain region based on automated anatomical labeling (AAL) atlas as feature representation from T1w MRI and rs-fMRI data of each subject, respectively. Then, in order to obtain features that are more helpful in identifying EMCI, a common multi-task feature selection method is applied. Afterwards, we construct a non-fully labelled subject graph using imaging and non-imaging phenotypic measures of each subject. Finally, a GCN model is adopted to perform the EMCI identification task.ResultsOur proposed EMCI identification method is evaluated on 210 subjects, including 105 subjects with EMCI and 105 normal controls (NCs), with both T1w MRI and rs-fMRI data from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database. Experimental results show that our proposed framework achieves an accuracy of 84.1% and an area under the receiver operating characteristic (ROC) curve (AUC) of 0.856 for EMCI/NC classification. In addition, by comparison, the accuracy and AUC values of our proposed framework are better than those of some existing methods in EMCI identification.ConclusionOur proposed EMCI identification framework is effective and promising for automatic diagnosis of EMCI in clinical practice.

Patterns of Regional Cerebral Blood Flow as a Function of Obesity in Adults.

Background:While obesity has been shown to be a risk factor for Alzheimer’s disease, the potential mechanisms underlying this risk may be clarified with better understanding of underlying physiology in obese persons.Objective:To identify patterns of cerebral perfusion abnormality in adults as a function of body mass index (BMI) defined weight categories, including overweight or obese status.Methods:A large psychiatric cohort of 35,442 brain scans across 17,721 adults (mean age 40.8±16.2 years, range 18–94 years) were imaged with SPECT during baseline and concentration scans, the latter done after each participant completed the Connors Continuous Performance Test II. ANOVA was done to identify patterns of perfusion abnormality in this cohort across BMI designations of underweight (BMI < 18.5), normal weight (BMI = 18.5 to 24.9), overweight (BMI 24.9 to 29.9), obesity (BMI≥30), and morbid obesity (BMI≥40). This analysis was done for 128 brain regions quantifying SPECT perfusion using the automated anatomical labeling (AAL) atlas.Results:Across adulthood, higher BMI correlated with decreased perfusion on both resting and concentration brain SPECT scans. These are seen in virtually all brain regions, including those influenced by AD pathology such as the hippocampus.Conclusion:Greater BMI is associated with cerebral perfusion decreases in both resting and concentration SPECT scans across adulthood.

Age-related differences in myeloarchitecture measured at 7 T

We have used the magnetisation transfer (MT) MRI measure as a primary measure of myelination in both the gray matter (GM) of the 78 cortical automated anatomical labeling (AAL) regions of the brain, and the underlying white matter in each region, in a cohort of healthy adults (aged 19–62 year old). The results revealed a significant quadratic trend in myelination with age, with average global myelination peaking at 42.9 year old in gray matter, and at 41.7 year old in white matter. We also explored the possibility of using the Nuclear Overhauser Enhancement (NOE) effect, which is acquired in a similar method to MT, as an additional measure of myelination. We found that the MT and NOE signals were strongly correlated in the brain and that the NOE effects displayed similar (albeit weaker) parabolic trends with age. We also investigated differences in cortical thickness with age, and confirmed a previous result of a linear decline of 4.5 ± 1.2 μm/y.

Identification of Alzheimer's disease based on wavelet transformation energy feature of the structural MRI image and NN classifier.

Alzheimer's disease (AD) is now difficult to be identified for clinicians, especially, at its prodromal stage, mild cognitive impairment (MCI), because of no obvious clinical symptom and few impacts on daily life at this phase. In addition, energy distribution differences of brain atrophies reflected in structural magnetic resonance imaging (sMRI) images between MCI patients and older healthy controls (HC) are minimal and subtle, which are difficult to be captured by the spatial analysis. In this study, we propose a novel method (namely AD-WTEF) to identify AD and MCI patients from HC subjects by extracting the wavelet transformation energy feature (WTEF) of the sMRI image. AD-WTEF firstly transforms each scan of the preprocessed sMRI image by wavelet to obtain its directional subbands with the same size at different transformation levels. And then, based on the anatomical automatic labeling (AAL) atlas, AD-WTEF constructs a new brain mask to segment the subbands at the same direction and transformation level into different energy regions of interest (EROIs). Thirdly, by averaging coefficients in an EROI, AD-WTEF gets an energy feature, following that energy features of different EROIs are connected to form an energy feature vector for describing the subbands at the same direction and transformation level. As a result, these energy feature vectors are further concatenated to be a WTEF of the sMRI image. Finally, the nearest neighbor (NN) classifier is selected and used for AD identification. Compared with other seven state-of-the-art methods, our AD-WTEF can effectively identify AD patients using the subtle energy distribution differences of sMRI images. Furthermore, experimental results indicate that our AD-WTEF can also find important brain ROIs related to AD.

Machine Learning Detects Pattern of Differences in Functional Magnetic Resonance Imaging (fMRI) Data between Chronic Fatigue Syndrome (CFS) and Gulf War Illness (GWI).

Background: Gulf War Illness (GWI) and Chronic Fatigue Syndrome (CFS) are two debilitating disorders that share similar symptoms of chronic pain, fatigue, and exertional exhaustion after exercise. Many physicians continue to believe that both are psychosomatic disorders and to date no underlying etiology has been discovered. As such, uncovering objective biomarkers is important to lend credibility to criteria for diagnosis and to help differentiate the two disorders. Methods: We assessed cognitive differences in 80 subjects with GWI and 38 with CFS by comparing corresponding fMRI scans during 2-back working memory tasks before and after exercise to model brain activation during normal activity and after exertional exhaustion, respectively. Voxels were grouped by the count of total activity into the Automated Anatomical Labeling (AAL) atlas and used in an “ensemble” series of machine learning algorithms to assess if a multi-regional pattern of differences in the fMRI scans could be detected. Results: A K-Nearest Neighbor (70%/81%), Linear Support Vector Machine (SVM) (70%/77%), Decision Tree (82%/82%), Random Forest (77%/78%), AdaBoost (69%/81%), Naïve Bayes (74%/78%), Quadratic Discriminant Analysis (QDA) (73%/75%), Logistic Regression model (82%/82%), and Neural Net (76%/77%) were able to differentiate CFS from GWI before and after exercise with an average of 75% accuracy in predictions across all models before exercise and 79% after exercise. An iterative feature selection and removal process based on Recursive Feature Elimination (RFE) and Random Forest importance selected 30 regions before exercise and 33 regions after exercise that differentiated CFS from GWI across all models, and produced the ultimate best accuracies of 82% before exercise and 82% after exercise by Logistic Regression or Decision Tree by a single model, and 100% before and after exercise when selected by any six or more models. Differential activation on both days included the right anterior insula, left putamen, and bilateral orbital frontal, ventrolateral prefrontal cortex, superior, inferior, and precuneus (medial) parietal, and lateral temporal regions. Day 2 had the cerebellum, left supplementary motor area and bilateral pre- and post-central gyri. Changes between days included the right Rolandic operculum switching to the left on Day 2, and the bilateral midcingulum switching to the left anterior cingulum. Conclusion: We concluded that CFS and GWI are significantly differentiable using a pattern of fMRI activity based on an ensemble machine learning model.

Brain functional network alterations caused by a strong desire to void in healthy adults: a graph theory analysis study.

This resting-state functional magnetic resonance imaging (fMRI) study determined the functional connectivity (FC) changes and topologic property alterations of the brain functional network provoked by a strong desire to void in healthy adults using a graph theory analysis (GTA). Thirty-four healthy, right-handed subjects filled their bladders by drinking water. The subjects were scanned under an empty bladder and a strong desire to void states. The Pearson's correlation coefficients were calculated among 90 brain regions in the automated anatomical labeling (AAL) atlas to construct the brain functional network. A paired t test (P < .05, after false discovery rate [FDR] correction) was used to detect significant differences in the FC, topologic properties (small-world parameters [gamma, sigma], Cp, Lp, Eglob, Eloc, and Enodal) between the two states in all subjects. Both the two states showed small-world network properties. The clustering coefficient (Cp) and local efficiency (Eloc) in the whole brain network decreased, while the FC within the default mode network (DMN) increased during the strong desire to void compared with the empty bladder state. Moreover, an increased nodal efficiency (Enodal) was detected in the basal ganglia (BG), DMN, sensorimotor-related network (SMN), and visual network (VN). We detected FC changes and topologic property alterations in brain functional networks caused by a strong desire to void in healthy and suggest that the micturition control may be a process dominated by DMN and coordinated by multiple sub-networks (such as, BG, SMN, and VN), which could serve as a baseline for understanding the pathologic process underlying bladder dysfunction and be useful to improve targeted therapy in the future.

Logistic Regression Algorithm Differentiates Gulf War Illness (GWI) Functional Magnetic Resonance Imaging (fMRI) Data from a Sedentary Control.

Gulf War Illness (GWI) is a debilitating condition characterized by dysfunction of cognition, pain, fatigue, sleep, and diverse somatic symptoms with no known underlying pathology. As such, uncovering objective biomarkers such as differential regions of activity within a Functional Magnetic Resonance Imaging (fMRI) scan is important to enhance validity of the criteria for diagnosis. Symptoms are exacerbated by mild activity, and exertional exhaustion is a key complaint amongst sufferers. We modeled this exertional exhaustion by having GWI (n = 80) and sedentary control (n = 31) subjects perform submaximal exercise stress tests on two consecutive days. Cognitive differences were assessed by comparing fMRI scans performed during 2-Back working memory tasks before and after the exercise. Machine learning algorithms were used to identify differences in brain activation patterns between the two groups on Day 1 (before exercise) and Day 2 (after exercise). The numbers of voxels with t > 3.17 (corresponding to p < 0.001 uncorrected) were determined for brain regions defined by the Automated Anatomical Labeling (AAL) atlas. Data were divided 70:30 into training and test sets. Recursive feature selection identified twenty-nine regions of interest (ROIs) that significantly distinguished GWI from control on Day 1 and 28 ROIs on Day 2. Ten regions were present in both models between the two days, including right anterior insula, orbital frontal cortex, thalamus, bilateral temporal poles, and left supramarginal gyrus and cerebellar Crus 1. The models had 70% accuracy before exercise on Day 1 and 85% accuracy after exercise on Day 2, indicating the logistic regression model significantly differentiated subjects with GWI from the sedentary control group. Exercise caused changes in these patterns that may indicate the cognitive differences caused by exertional exhaustion. A second set of predictive models was able to classify previously identified GWI exercise subgroups START, STOPP, and POTS for both Days 1 and Days 2 with 67% and 69% accuracy respectively. This study was the first of its kind to differentiate GWI and the three sub-phenotypes START, STOPP, and POTS from a sedentary control using a logistic regression estimation method.

Impact of the size of the normal database on the performance of the specific binding ratio in dopamine transporter SPECT

BackgroundThis study investigated the impact of the size of the normal database on the classification performance of the specific binding ratio (SBR) in dopamine transporter (DAT) SPECT with [123I]FP-CIT in different settings.MethodsThe first subject sample comprised 645 subjects from the Parkinson’s Progression Marker Initiative (PPMI), 207 healthy controls (HC), and 438 Parkinson’s disease (PD) patients. The second sample comprised 372 patients from clinical routine patient care, 186 with non-neurodegenerative parkinsonian syndrome (PS) and 186 with neurodegenerative PS. Single-photon emission computed tomography (SPECT) images of the clinical sample were reconstructed with two different reconstruction algorithms (filtered backprojection, iterative ordered subsets expectation maximization (OSEM) reconstruction with resolution recovery). The putaminal specific binding ratio (SBR) was computed using an anatomical region of interest (ROI) predefined in standard (MNI) space in the Automated Anatomic Labeling (AAL) atlas or using hottest voxels (HV) analysis in large predefined ROIs. SBR values were transformed to z-scores using mean and standard deviation of the SBR in a normal database of varying sizes (n = 5, 10, 15,…, 50) randomly selected from the HC subjects (PPMI sample) or the patients with non-neurodegenerative PS (clinical sample). Accuracy, sensitivity, and specificity for identifying patients with PD or neurodegenerative PS were determined as performance measures using a predefined fixed cutoff on the z-score. This was repeated for 10,000 randomly selected normal databases, separately for each size of the normal database. Mean and 5th percentile of the performance measures over the 10,000 realizations were computed. Accuracy, sensitivity, and specificity when using the whole set of HC or non-neurodegenerative PS subjects as normal database were used as benchmark.ResultsMean loss of accuracy of the putamen SBR z-score was below 1% when the normal database included at least 15 subjects, independent of subject sample (PPMI or clinical), reconstruction method (filtered backprojection or OSEM), and ROI method (AAL or HV). However, the variability of the accuracy of the putamen SBR z-score decreased monotonically with increasing size of normal database and was still considerable at size 15. In order to achieve less than 5% “maximum” loss of accuracy (defined by the 5th percentile) in all settings required at least 25 to 30 subjects in the normal database. Reduction of mean and “maximum” loss of accuracy of the putamen SBR z-score by further increasing the size of the normal database was very small beyond size 40.ConclusionsThe results of this study suggest that 25 to 30 is the minimum size of the normal database to reliably achieve good performance of semi-quantitative analysis in dopamine transporter (DAT) SPECT, independent of the algorithm used for image reconstruction and the ROI method used to estimate the putaminal SBR.

Stage detection of mild cognitive impairment via fMRI using Hilbert Huang transform based classification framework.

This work aims to establish a classification framework for the diagnosis of mild cognitive impairment (MCI) at different stages (early MCI and late MCI) through direct analysis of resting-state functional magnetic resonance imaging (rs-fMRI) signals and using the accuracy (total correct rate), specificity (correct rate of late MCI) and sensitivity (correct rate of early MCI) to validate its classification performance. All fMR images of subjects were parcellated into 116 regions of interest (ROIs) by applying the Anatomical Automatic Labeling (AAL) template, and the average rs-fMRI signals of each ROI were extracted. The Hilbert-Huang transform (HHT) was introduced into the framework to decompose each rs-fMRI signal into a series of intrinsic mode functions (IMFs) and to analyze these nonstationary and nonlinear time-series from the perspective of multiresolution. After obtaining the instantaneous frequencies and amplitudes of all IMFs of a signal, the Hilbert weighted frequencies (HWFs) were calculated and combined into a vector as the feature of the corresponding ROI. Support Vector Machine (SVM) was implemented to classify MCI at different stages. We used the independent two-sample t-test as the feature selection method and measured the classification performance through the leave-one-out cross-validation (LOOCV) method. Results on 77 early MCI (eMCI) and 64 late MCI (lMCI) with baseline rs-fMRI data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) yielded 87.94% classification accuracy. Some of the brain regions with significant differences found by previous studies have been confirmed in this work. We found that HWF characteristics exhibited a significant downward trend in all cerebellar regions. The rs-fMRI signals in differential brain regions have not changed completely, but only altered in some narrow frequency bands. The analysis results showed that during the progress of MCI, the main changes of rs-fMRI were concentrated in IMF3, while IMFs with other indexes also contained HWF features with high SVM weights, such as Orbitofrontal superior frontal gyrus in IMF2, Insula in IMF4, and Lobule Ⅲ of vermis in IMF5, indicating that other IMFs provide important information for the diagnosis of MCI as well. This work confirmed the classification ability of HHT-based classification framework in classification of at different stages of MCI. Through the analysis, we found that during the progress of MCI the main changes of rs-fMRI were concentrated in IMF3, and HWF characteristics showed a significant downward trend in all cerebellar regions.

Abnormal Rich Club Organization of Structural Network as a Neuroimaging Feature in Relation With the Severity of Primary Insomnia.

PurposeInsomnia is the most prevalent sleep complaint in the general population but is often intractable due to uncertainty regarding the underlying pathomechanisms. Sleep is regulated by a network of neural structures interconnected with the core nodes of the brain connectome referred to as the “rich club”. We examined alterations in brain rich-club organization as revealed by diffusion tensor imaging (DTI) and the statistical relationships between abnormalities in rich-club metrics and the clinical features of primary insomnia (PI).MethodsThis study recruited 43 primary insomnia (PI) patients and 42 age-, sex-, and education level-matched healthy controls (HCs). Differences in global and regional network parameters between PI and healthy control groups were compared by nonparametric tests, and Spearman's correlations were calculated to assess associations of these network metrics with PI-related clinical features, including disease duration and scores on the Pittsburgh Sleep Quality Index, Insomnia Severity Index, Self-Rating Anxiety Scale, and Self-Rating Depression Scale.ResultsWeighted white matter networks exhibited weaker rich-club organization in PI patients than HCs across different thresholds (50%, 75%, and 90%) and parcellation schemes [automated anatomical labeling (AAL)-90 and AAL-1024]. Aberrant rich-club organization was found mainly in limbic-cortical-basal ganglia circuits and the default-mode network.ConclusionsAbnormal rich-club metrics are a characteristic feature of PI-related to disease severity. These metrics provide potential clues to PI pathogenesis and may be useful as diagnostic markers and for assessment of treatment response.

Altered resting-state functional networks in patients with hemodialysis: a graph-theoretical based study.

Recent studies have demonstrated that hemodialysis patients exhibit disruptions in functional networks with invisible cerebral alterations. We explored the alterations of functional connectivity in hemodialysis patients using the graph-theory method. A total of 46 hemodialysis patients (53.11 ± 1.58 years, 28 males) and 47 healthy controls (55.57 ± 0.86 years, 22 males) were scanned byusing resting-state functional magnetic resonance imaging. The brains of these patients were divided into 90 regions and functional connectivity was constructed with the automatic anatomical labeling atlas. In the defined threshold range, the graph-theory analysis was performed to compare the topological properties including global, regional and edge parameters between the hemodialysis and the healthy control groups. Both hemodialysis patients and healthy control subjects demonstrated common small-world property of the brain functional connections. At the global level, the parameters normalized clustering coefficients and small-worldness were significantly decreased in hemodialysis patients compared with those noted in healthy controls. At the regional level, abnormal nodal metrics (increased or decreased nodal degree, betweenness centrality and efficiency) were widely found in hemodialysis patients compared with those of healthy controls. The network-based statistical method was employed and two disrupted neural circuits with 18 nodes and 19 edges (P = 0.0139, corrected) and 10 nodes and 11 edges (P = 0.0399, corrected) were detected. Of note, the edge-increased functional connectivity was associated with the salience network and the frontal-temporal-basal ganglia connection, whereas the edge-decreased functional connectivity was associated with the frontoparietal network. The graph-theory method may be one of the potential tools to detect disruptions of cerebral functional connectivity and provide important evidence for understanding the neuropathology of hemodialysis patients from the disrupted network organization perspective.

Juvenile myoclonic epilepsy shows increased posterior theta, and reduced sensorimotor beta resting connectivity

BackgroundWidespread structural and functional brain network changes have been shown in Juvenile Myoclonic Epilepsy (JME) despite normal clinical neuroimaging. We sought to better define these changes using magnetoencephalography (MEG) and source space connectivity analysis for optimal neurophysiological and anatomical localisation. MethodsWe consecutively recruited 26 patients with JME who underwent resting state MEG recording, along with 26 age-and-sex matched controls. Whole brain connectivity was determined through correlation of Automated Anatomical Labelling (AAL) atlas source space MEG timeseries in conventional frequency bands of interest delta (1−4 Hz), theta (4−8 Hz), alpha (8−13 Hz), beta (13−30 Hz) and gamma (40−60 Hz). We used a Linearly Constrained Minimum Variance (LCMV) beamformer to extract voxel wise time series of ‘virtual sensors’ for the desired frequency bands, followed by connectivity analysis using correlation between frequency- and node-specific power fluctuations, for the voxel maxima in each AAL atlas label, correcting for noise, potentially spurious connections and multiple comparisons. ResultsWe found increased connectivity in the theta band in posterior brain regions, surviving statistical correction for multiple comparisons (corrected p < 0.05), and decreased connectivity in the beta band in sensorimotor cortex, between right pre- and post- central gyrus (p < 0.05) in JME compared to controls. ConclusionsAltered resting-state MEG connectivity in JME comprised increased connectivity in posterior theta – the frequency band associated with long range connections affecting attention and arousal - and decreased beta-band sensorimotor connectivity. These findings likely relate to altered regulation of the sensorimotor network and seizure prone states in JME.

Hemangiopericytomas: Spatial Intracranial Location in a Voxel-Based Mapping Study.

To investigate the preferred location of intracranial hemangiopericytomas (IHPCs) with voxel-based mapping and 3-dimensional reconstruction from MRI data. Gadolinium-enhanced tumors of 258 primary and single IHPCs were segmented semi-automatically, followed by manual checking and editing of boundaries. The lesions were registered to Montreal Neurological Institute standard anatomical space, and heat-map and 3-dimensional rendered frequency images were generated. All tumors were then superimposed on the Anatomical Automatic Labeling (AAL) template to further investigate the difference in the tumor location based on the voxel-wise frequency of occurrence with respect to laterality, sex, age, and pathologic grade. The 3-dimensional rendered images show that the tumors commonly located in the posterior cranial cavity, surrounding the tentorium. The posterior third of the superior sagittal sinus and the confluence of sinuses were commonly affected. According to the analysis of tumor occurrence frequency in the AAL template, IHPCs were mainly observed in the limbic lobe, occipital lobe, and cerebellum. Tumors in younger patients preferentially located in the right occipital region (P = .027), whereas those with higher pathological grade more often located in the left parietal lobe (P = .034). This is the first voxel-based study to explore the predilection site of IHPCs. Our study suggests that these tumors commonly affect the posterior cranial cavity, adjoining the tentorium and venous sinus. Further research is needed to investigate the possible factors underlying these topographic preferences.

Enhancing the feature representation of multi-modal MRI data by combining multi-view information for MCI classification

The classification of mild cognitive impairment (MCI), which is a early stage of Alzheimer’s disease and is associated with brain structural and functional changes, is still a challenging task. Recent studies have shown great promise for improving the performance of MCI classification by combining multiple structural and functional features, such as grey matter volume and clustering coefficient. However, extracting which features and how to combine multiple features to improve the performance of MCI classification have always been challenging problems. To address these problems, in this study we propose a new method to enhance the feature representation of multi-modal MRI data by combining multi-view information to improve the performance of MCI classification. Firstly, we extract two structural features (including grey matter volume and cortical thickness) and two functional features (including clustering coefficient and shortest path length) of each cortical brain region based on automated anatomical labeling (AAL) atlas from both T1w MRI and rs-fMRI data of each subject. Then, in order to obtain features that are more helpful in distinguishing MCI subjects, an improved multi-task feature selection method, namely MTFS-gLASSO-TTR, is proposed. Finally, a multi-kernel learning algorithm is adopted to combine multiple features to perform the MCI classification task. Our proposed MCI classification method is evaluated on 315 subjects (including 105 LMCI subjects, 105 EMCI subjects and 105 NCs) with both T1w MRI and rs-fMRI data from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database. Experimental results show that our proposed method achieves an accuracy of 88.5% and an area under the receiver operating characteristic (ROC) curve (AUC) of 0.897 for LMCI/NC classification, an accuracy of 82.7% and an AUC of 0.832 for EMCI/NC classification, and an accuracy of 79.6% and an AUC of 0.803 for LMCI/EMCI classification, respectively. In addition, by comparison, the accuracy and AUC values of our proposed method are better than those of some existing state-of-the-art methods in MCI classification. Overall, our proposed MCI classification method is effective and promising for automatic diagnosis of MCI in clinical practice.

A Machine Learning Approach to the Differentiation of Functional Magnetic Resonance Imaging Data of Chronic Fatigue Syndrome (CFS) From a Sedentary Control.

Chronic Fatigue Syndrome (CFS) is a debilitating condition estimated to impact at least 1 million individuals in the United States, however there persists controversy about its existence. Machine learning algorithms have become a powerful methodology for evaluating multi-regional areas of fMRI activation that can classify disease phenotype from sedentary control. Uncovering objective biomarkers such as an fMRI pattern is important for lending credibility to diagnosis of CFS. fMRI scans were evaluated for 69 patients (38 CFS and 31 Control) taken before (Day 1) and after (Day 2) a submaximal exercise test while undergoing the n-back memory paradigm. A predictive model was created by grouping fMRI voxels into the Automated Anatomical Labeling (AAL) atlas, splitting the data into a training and testing dataset, and feeding these inputs into a logistic regression to evaluate differences between CFS and control. Model results were cross-validated 10 times to ensure accuracy. Model results were able to differentiate CFS from sedentary controls at a 80% accuracy on Day 1 and 76% accuracy on Day 2 (Table 3). Recursive features selection identified 29 ROI's that significantly distinguished CFS from control on Day 1 and 28 ROI's on Day 2 with 10 regions of overlap shared with Day 1 (Figure 3). These 10 shared regions included the putamen, inferior frontal gyrus, orbital (F3O), supramarginal gyrus (SMG), temporal pole; superior temporal gyrus (T1P) and caudate ROIs. This study was able to uncover a pattern of activated neurological regions that differentiated CFS from Control. This pattern provides a first step toward developing fMRI as a diagnostic biomarker and suggests this methodology could be emulated for other disorders. We concluded that a logistic regression model performed on fMRI data significantly differentiated CFS from Control.