Biology subjects

Calhoun, V.

Publications and source records attributed to Calhoun, V..

At least 145 records · Page 8Linked to original sources

bioRxiv2021

Multi-Spatial Scale Dynamic Interactions between Functional Sources Reveal Sex-Specific Changes in Schizophrenia

We introduce an extension of independent component analysis (ICA), called multiscale ICA (msICA), and design an approach to capture dynamic functional source interactions within and between multiple spatial scales. msICA estimates functional sources at multiple spatial scales without imposing direct constraints on the size of functional sources, overcomes the limitation of using fixed anatomical locations, and eliminates the need for model-order selection in ICA analysis. We leveraged this approach to study sex-specific and -common connectivity patterns in schizophrenia. Results show dynamic reconfiguration and interaction within and between multi-spatial scales. Sex-specific differences occur (1) within the subcortical domain, (2) between the somatomotor and cerebellum domains, and (3) between the temporal domain and several others, including the subcortical, visual, and default mode domains. Most of the sex-specific differences belong to between-spatial scale functional interactions and are associated with a dynamic state with strong functional interactions between the visual, somatomotor, and temporal domains and their anticorrelation patterns with the rest of the brain. We observed significant correlations between multi-spatial scale functional interactions and symptom scores, highlighting the importance of multiscale analyses to identify potential biomarkers for schizophrenia. As such, we recommend such analyses as an important option for future functional connectivity studies.

neuroscience↗

bioRxiv2021

Disrupted dynamic functional network connectivity among cognitive control networks in the progression of Alzheimer's disease

BackgroundAlzheimers disease (AD) is the most common age-related dementia that promotes a decline in memory, thinking, and social skills. The initial stages of dementia can be associated with mild symptoms, and symptom progression to a more severe state is heterogeneous across patients. Recent work has demonstrated the potential for functional network mapping to assist in the prediction of symptomatic progression. However, this work has primarily used static functional connectivity (sFC) from rs-fMRI. Recently, dynamic functional connectivity (dFC) has been recognized as a powerful advance in functional connectivity methodology to differentiate brain network dynamics between healthy and diseased populations. MethodsGroup independent component analysis was applied to extract 17 components within the cognitive control network (CCN) from 1385 individuals across varying stages of AD symptomology. We estimated dFC among 17 components within the CCN, followed by clustering the dFCs into 3 recurring brain states and then estimated a hidden Markov model and the occupancy rate for each subject. Finally, we investigated the link between CCN dFC connectivity features with AD progression. ResultsProgression of AD symptoms were associated with increases in connectivity within the middle frontal gyrus. Also, the AD with mild and severer symptoms showed less connectivity within the inferior parietal lobule and between this region with the rest of CCN. Finally, comparing with mild dementia, we found that the normal brain spends significantly more time in a state with lower within middle frontal gyrus connectivity and higher connectivity between the hippocampus and the rest of CCN, highlighting the importance of assessing the dynamics of brain connectivity in this disease. ConclusionOur results suggest that AD progress not only alters the CCN connectivity strength but also changes the temporal properties in this brain network. This suggests the temporal and spatial pattern of CCN as a biomarker that differentiates different stages of AD. Impact StatementBy assuming that functional connectivity is static over time, many of previous studies have ignored the brain dynamic in Alzheimers disease progression. Here, a longitudinal resting-state functional magnetic resonance imaging data are used to explore the temporal changes of functional connectivity in the cognitive control network in Alzheimers disease progression. The result of this study would increase our understanding about the underlying mechanisms of Alzheimers Disease and help in finding future treatment of this neurological disorder.

neuroscience↗

bioRxiv2020

Building Models of Functional Interactions Among Brain Domains that Encode Varying Information Complexity: A Schizophrenia Case Study

Revealing associations among various structural and functional patterns of the brain can yield highly informative results about the healthy and disordered brain. Studies using neuroimaging data have more recently begun to utilize the information within as well as across various functional and anatomical domains (i.e., groups of brain networks). However, most whole-brain approaches assume similar complexity of interactions throughout the brain. Here we investigate the hypothesis that interactions between brain networks capture varying amounts of complexity, and that we can better capture this information by varying the complexity of the model subspace structure based on available training data. To do this, we employ a Bayesian optimization-based framework known as the Tree Parzen Estimator (TPE) to identify, exploit and analyze patterns of variation in the information encoded by temporal information extracted from functional magnetic resonance imaging (fMRI) subdomains of the brain. Using a repeated cross-validation procedure on a schizophrenia classification task, we demonstrate evidence that interactions between specific functional subdomains are better characterized by more sophisticated model architectures compared to less complicated ones required by the others for optimally contributing towards classification and understanding the brains functional interactions. We show that functional subdomains known to be involved in schizophrenia require more complex architectures to optimally unravel discriminatory information about the disorder. Our study points to the need for adaptive, hierarchical learning frameworks that cater differently to the features from different subdomains, not only for a better prediction but also for enabling the identification of features predicting the outcome of interest.

neuroscience↗

bioRxiv2020

Impact of Model Order Choice on the Results of Parallel Independent Component Analysis

Parallel independent component analysis (pICA) is a data-driven method that identifies the maximally independent components of multiple imaging modalities while simultaneously investigating the strength of their correlations. Researchers using pICA are given the option to use the suggested model order calculated by the minimum descriptive length (MDL) algorithm, or they can choose their own model order. To date, there are no suggested guidelines for this choice. To test the sensitivity of pICA to the selection of model order, we applied it to a well-researched brain disorder, schizophrenia, looking at the correlations between patterns of grey matter volume (GM) volume and white matter integrity, measured using fractional anisotropy (FA). We varied model orders from low to high, and tested the sensitivity to disorder effects (cases vs controls), similarity of spatial maps identified across model orders, consolidation or distribution effects related to model order selection, and the performance of the minimum descriptive length (MDL) algorithm. The pICA results (multimodal analysis) were also compared to the ICA (unimodal analysis) for each imaging modality. Across model orders, there was consistent sensitivity to disorder effects, and clustered patterns of spatial maps for both the GM and FA reflecting those differences. The MDL-estimated model order captured the majority, but not all, of the spatial patterns present in the GM and FA. There was not the expected consolidation of spatial maps at lower model orders, nor the distribution of spatial maps at higher model orders. The spatial patterns identified in the ICA closely resemble those found in the pICA, although lacking the benefit of the optimization algorithm, were not as highly correlated. This offers some insight and guidance for researchers interested in using pICA with regard to selecting model order for their particular analysis of multiple imaging modalities.

neuroscience↗

bioRxiv2020

Statelets: high dimensional predominant shapes in dynamic functional network connectivity

Dynamic functional network connectivity (dFNC) analysis is a widely used approach for capturing brain activation patterns, connectivity states, and network organization. However, a typical sliding window plus clustering (SWC) approaches for analyzing dFNC continuously models the system through a fixed set of connectivity patterns or states. It assumes these patterns are span throughout the brain, but in practice, they are more spatially constrained and temporally short-lived. Thus, SWC is not designed to capture transient dynamic changes nor heterogeneity across subjects/time. Here, we adapt time series motifs to model the temporal dynamics of functional connectivity. We propose a state-space data mining approach that combines a probabilistic pattern summarization framework called Statelets -- a subset of high dimensional state-shape prototypes capturing the dynamics. We handle scale differences using the earth mover distance and utilize kernel density estimation to build a probability density profile for local motifs. We apply the framework to study dFNC collected from patients with schizophrenia (SZ) and healthy control (HC). Results demonstrate SZ subjects exhibit reduced modularity in their brain network organization relative to HC. These statelets in the HC group show more recurrence across the dFNC time-course compared to the SZ. An analysis of the consistency of the connections across time reveals significant differences within visual, sensorimotor, and default mode regions where HC subjects show higher consistency than SZ. The introduced statelet-approach also enables the handling of dynamic information in cross-modal applications to study healthy and disordered brains and multi-modal fusion within a single dataset.

neuroscience↗

bioRxiv2020

Detection of Prenatal Alcohol Exposure Using Machine Learning Classification of Resting-State Functional Network Connectivity Data

IntroductionPrevious work utilizing resting state fMRI to measure functional network connectivity in rodents with moderate prenatal alcohol exposure (PAE) revealed several sex- and region-dependent alterations in FNC implicating FNC as potential biomarker for PAE. Given that FNC is sensitive to neurological and psychiatric conditions in humans, here, we explore the use of previously acquired FNC data and machine learning methods to detect PAE among a sample of rodents exposed to moderate PAE and controls exposed to a saccharin solution. Materials & MethodsWe utilized previously acquired fMRI data from 48 adult rats 24 PAE (12 male 12 female) and 24 saccharin exposed (SAC) controls (12 male and 12 female) for classification. The entire data sample was utilized to perform binary classification (SAC or PAE) of FNC data with multiple support vector machine (SVM) kernels and out-of-sample cross-validation to evaluate classification performance. ResultsResults revealed accuracy rates of 62.5% for all samples, 58.3% for males, and 79.2% for females using a quadratic SVM kernel to classify moderate PAE from FNC data. In addition, brain networks localized to hippocampal and cortical regions contributed strongly to QSVM classifications. ConclusionOur results suggest overall modest classification performance of a QSVM to detect moderate PAE from FNC data gathered from adult rats, yet good performance among females. Further developments and refinement of the technique hold promise for the detection of PAE in earlier developmental time periods that potentially offer additional tools for the non-invasive detection of PAE from FNC data. IMPACT STATEMENTThe diagnosis of fetal alcohol spectrum disorders (FASDs) can be challenging in individuals who lack the hallmark facial dysmorphologies associated with heavy prenatal alcohol exposure (PAE). The absence of a diagnosis prevents individuals with PAE from receiving the treatment and services that improves quality of life outcomes. This research explores the use of preclinical functional network connectivity data and machine learning techniques as a novel and non-invasive means of detecting PAE. Our aim is to contribute basic science towards improving diagnostic strategies that can lead to securing timely and appropriate support for individuals with FASD and their caregivers.

neuroscience↗

bioRxiv2020

Modular and state-relevant connectivity in high-frequency resting-state BOLD fMRI data: An independent component analysis

Resting-state fMRI (rs-fMRI) data are typically filtered at different frequency bins between 0.008[~]0.2 Hz (varies across the literature) prior to analysis to mitigate nuisance variables (e.g., drift, motion, cardiac, and respiratory) and maximize the sensitivity to neuronal-mediated BOLD signal. However, multiple lines of evidence suggest meaningful BOLD signal may also be parsed at higher frequencies. To test this notion, a functional network connectivity (FNC) analysis based on a spatially informed independent component analysis (ICA) was performed at seven different bandpass frequency bins to examine FNC matrices across spectra. Further, eyes open (EO) vs. eyes closed (EC) resting-state acquisitions from the same participants were compared across frequency bins to examine if EO vs. EC FNC matrices and randomness estimations of FNC matrices are distinguishable at different frequencies. Results show that FNCs in higher-frequency bins display modular FNC similar to the lowest frequency bin, while r-to-z FNC and FNC-based measures indicating matrix non-randomness were highest in the 0.31-0.46 Hz range relative to all frequency bins above and below this range. As such, the FNC within this range appears to be the most temporally correlated, but the mechanisms facilitating this coherence require further analyses. Compared to EO, EC displayed greater FNC (involved in visual, cognitive control, somatomotor, and auditory domains) and randomness values at lower frequency bins, but this phenomenon flipped (EO > EC) at frequency bins greater than 0.46 Hz, particularly within visual regions. While the effect sizes range from small to large specific to frequency range and resting state (EO vs. EC), with little influence from common artifacts. These differences indicate that unique information can be derived from FNC between BOLD signals at different frequencies relative to a given restingstate acquisition and support the hypothesis meaningful BOLD signal is present at higher frequency ranges.

neuroscience↗

bioRxiv2020

A classification-based approach to estimate the number of resting fMRI dynamic functional connectivity states

Recent work has focused on the study of dynamic (vs static) brain connectivity in resting fMRI data. In this work, we focus on temporal correlation between time courses extracted from coherent networks or components called functional network connectivity (FNC). Dynamic functional network connectivity (dFNC) is most commonly estimated using a sliding window-based approach to capture short periods of FNC change. These data are then clustered to estimate transient connectivity patterns or states. Determining the number of states is a challenging problem. The elbow criterion is a widely used approach to determine the optimal number of states. In our work, we present an alternative approach that evaluates classification (e.g. healthy controls versus patients) as a measure to select the optimal number of states (clusters). We apply different classification strategies to perform classification between healthy controls (HC) and patients with schizophrenia (SZ) for different numbers of states (i.e. varying the model order in the clustering algorithm). We compute cross-validated accuracy for different model orders to evaluate the classification performance. Our results are consistent with our earlier work which shows that overall accuracy improves when dynamic connectivity measures are used separately or in combination with static connectivity measures. Results also show that the optimal model order for classification is different from that using the standard k-means model selection method and that such optimization improves resulting in cross-validated accuracy. The optimal model order obtained from the proposed approach also gives significantly improved classification performance over the traditional model selection method. In sum, the observed results suggest that if ones goal is to perform classification, using the proposed approach as a criterion for selecting the optimal number of states in dynamic connectivity analysis leads to improved accuracy in hold-out data.

neuroscience↗

bioRxiv2020

Addressing Inaccurate Nosology in Mental Health: A Multi Label Data Cleansing Approach for Detecting Label Noise from Structural Magnetic Resonance Imaging Data in Mood and Psychosis Disorders

BackgroundMental health diagnostic approaches are seeking to identify biological markers to work alongside advanced machine learning approaches. It is difficult to identify a biological marker of disease when the traditional diagnostic labels themselves are not necessarily valid. MethodsWe worked with T1 structural magnetic resonance imaging data collected from individuals with mood and psychosis disorders from over 1400 individuals comprising healthy controls, psychosis patients and their unaffected first-degree relatives including 176 bipolar probands, 134 schizoaffective probands, 240 schizophrenia proband, 581 patients relatives and 362 controls. We assumed there might be noise in the diagnostic labeling process. We detected label noise by classifying the data multiple times using a support vector machine classifier, and then we flagged those individuals in which all classifiers unanimously mislabeled those subjects. Next, we assigned a new diagnostic label to these individuals, based on the biological data (MRI), using iterative data cleansing approach. ResultsSimulation results showed our method was highly accurate in identifying label noise. Both diagnostic and Biotype categories showed about 65% and 63% respectively of noisy labels with the largest amount of relabeling occurring between the healthy control and bipolar and schizophrenia disorder individuals as well as in the unaffected close relatives. The extraction of imaging features highlighted regional brain changes associated with each group. ConclusionsThis approach represents an initial step towards developing strategies that need not assume existing mental health diagnostic categories are always valid, but rather allows us to leverage this information while also acknowledging that there are misassignments.

neuroscience↗

bioRxiv2020

Cortical Thickness Trajectories across the Lifespan: Data from 17,075 healthy individuals aged 3-90 years

Delineating age-related cortical trajectories in healthy individuals is critical given the association of cortical thickness with cognition and behaviour. Previous research has shown that deriving robust estimates of age-related brain morphometric changes requires large-scale studies. In response, we conducted a large-scale analysis of cortical thickness in 17,075 individuals aged 3-90 years by pooling data through the Lifespan Working group of the Enhancing Neuroimaging Genetics through Meta-Analysis (ENIGMA) Consortium. We used fractional polynomial (FP) regression to characterize age-related trajectories in cortical thickness, and we computed normalized growth centiles using the parametric Lambda, Mu, and Sigma (LMS) method. Inter-individual variability was estimated using meta-analysis and one-way analysis of variance. Overall, cortical thickness peaked in childhood and had a steep decrease during the first 2-3 decades of life; thereafter, it showed a gradual monotonic decrease which was steeper in men than in women particularly in middle-life. Notable exceptions to this general pattern were entorhinal, temporopolar and anterior cingulate cortices. Inter-individual variability was largest in temporal and frontal regions across the lifespan. Age and its FP combinations explained up to 59% variance in cortical thickness. These results reconcile uncertainties about age-related trajectories of cortical thickness; the centile values provide estimates of normative variance in cortical thickness, and may assist in detecting abnormal deviations in cortical thickness, and associated behavioural, cognitive and clinical outcomes.

neuroscience↗

bioRxiv2020

Subcortical Volume Trajectories across the Lifespan: Data from 18,605 healthy individuals aged 3-90 years

Age has a major effect on brain volume. However, the normative studies available are constrained by small sample sizes, restricted age coverage and significant methodological variability. These limitations introduce inconsistencies and may obscure or distort the lifespan trajectories of brain morphometry. In response, we capitalised on the resources of the Enhancing Neuroimaging Genetics through Meta-Analysis (ENIGMA) Consortium to examine the age-related morphometric trajectories of the ventricles, the basal ganglia (caudate, putamen, pallidum, and nucleus accumbens), the thalamus, hippocampus and amygdala using magnetic resonance imaging data obtained from 18,605 individuals aged 3-90 years. All subcortical structure volumes were at their maximum early in life; the volume of the basal ganglia showed a gradual monotonic decline thereafter while the volumes of the thalamus, amygdala and the hippocampus remained largely stable (with some degree of decline in thalamus) until the sixth decade of life followed by a steep decline thereafter. The lateral ventricles showed a trajectory of continuous enlargement throughout the lifespan. Significant age-related increase in inter-individual variability was found for the hippocampus and amygdala and the lateral ventricles. These results were robust to potential confounders and could be used to derive risk predictions for the early identification of diverse clinical phenotypes.

neuroscience↗

bioRxiv2020

Hype versus hope: Deep learning encodes more predictive and robust brain imaging representations than standard machine learning

Previous successes of deep learning (DL) approaches on several complex tasks have hugely inflated expectations of their power to learn subtle properties of complex brain imaging data, and scale to large datasets. Perhaps as a reaction to this inflation, recent critical commentaries unfavorably compare DL with standard machine learning (SML) approaches for the analysis of brain imaging data. Yet, their conclusions are based on pre-engineered features which deprives DL of its main advantage: representation learning. Here we evaluate this and show the importance of representation learning for DL performance on brain imaging data. We report our findings from a large-scale systematic comparison of SML approaches versus DL profiled in a ten-way age and gender-based classification task on 12,314 structural MRI images. Results show that DL methods, if implemented and trained following the prevalent DL practices, have the potential to substantially improve compared to SML approaches. We also show that DL approaches scale particularly well presenting a lower asymptotic complexity in relative computational time, despite being more complex. Our analysis reveals that the performance improvement saturates as the training sample size grows, but shows significantly higher performance throughout. We also show evidence that the superior performance of DL is primarily due to the excellent representation learning capabilities and that SML methods can perform equally well when operating on representations produced by the trained DL models. Finally, we demonstrate that DL embeddings span a comprehensible projection spectrum and that DL consistently localizes discriminative brain biomarkers, providing an example of the robustness of prediction relevance estimates. Our findings highlight the presence of non-linearities in brain imaging data that DL frameworks can exploit to generate superior predictive representations for characterizing the human brain, even with currently available data sizes.

neuroscience↗

bioRxiv2020

Greater male than female variability in regional brain structure across the lifespan

For many traits, males show greater variability than females, with possible implications for understanding sex differences in health and disease. Here, the ENIGMA (Enhancing Neuro Imaging Genetics through Meta-Analysis) Consortium presents the largest-ever mega-analysis of sex differences in variability of brain structure, based on international data spanning nine decades of life. Subcortical volumes, cortical surface area and cortical thickness were assessed in MRI data of 16,683 healthy individuals 1-90 years old (47% females). We observed significant patterns of greater male than female between-subject variance for all subcortical volumetric measures, all cortical surface area measures, and 60% of cortical thickness measures. This pattern was stable across the lifespan for 50% of the subcortical structures, 70% of the regional area measures, and nearly all regions for thickness. Our findings that these sex differences are present in childhood implicate early life genetic or gene-environment interaction mechanisms. The findings highlight the importance of individual differences within the sexes, that may underpin sex-specific vulnerability to disorders.

neuroscience↗

bioRxiv2019

Decentralized multi-site VBM analysis during adolescence shows structural changes linked to age, body mass index, and smoking: A COINSTAC analysis

In the recent past, there has been an upward trend in developing frameworks that enable neuroimaging researchers to address challenging questions by leveraging data across multiple sites all over the world. One such framework, Collaborative Informatics and Neuroimaging Suite Toolkit for Anonymous Computation (COINSTAC), provides a platform to analyze neuroimaging data stored locally across multiple organizations without the need for pooling the data at any point during the analysis. In this paper, we perform a decentralized voxel-based morphometry analysis of structural magnetic resonance imaging data across two different sites to understand the structural changes in the brain as linked to age, body mass index and smoking. Results produced by the decentralized analysis are contrasted with similar findings in literature. This work showcases the potential benefits of performing multi-voxel and multivariate analyses of large-scale neuroimaging data located at multiple sites.

neuroscience↗

bioRxiv2019

Age-related Structural and Functional Variations in 5967 Individuals across the Adult Lifespan

Exploring brain changes across the human lifespan is becoming an important topic in neuroscience. Though there are multiple studies which investigated the relationship between age and brain imaging data, the results are heterogeneous due to small sample sizes and relatively narrow age ranges. Here, based on year-wise estimation of 5967 subjects from 13 to 72 years old, we aimed to provide a more precise description of adult lifespan variation trajectories of gray matter volume (GMV), structural network correlation (SNC) and functional network connectivity (FNC) using independent component analysis and multivariate linear regression model. Our results revealed the following relationships: 1)GMV linearly declined with age in most regions, while parahippocampus showed an inverted U-shape quadratic relationship with age; SNC presented a U-shape quadratic relationship with age within cerebellum, and inverted U-shape relationship primarily in the default mode network (DMN) and frontoparietal (FP) related correlation. 2) FNC tended to linearly decrease within resting-state networks (RSNs), especially in visual network and DMN. Early increase was revealed between RSNs, primarily in FP and DMN, which experienced decrease at older ages. U-shape relationship was also revealed to compensate for the cognition deficit in attention and subcortical related connectivity at late years. 3) The link between middle occipital gyrus and insula, as well as precuneus and cerebellum, exhibited similar changing trends between SNC and FNC across the adult lifespan. Collectively, these results highlight the benefit of lifespan study and provide a precise description of age-related regional variation and SNC/FNC changes based on a large dataset.

neuroscience↗

bioRxiv2019

Disambiguating the role of blood flow and global signal with Partial Information Decomposition

In resting state functional magnetic resonance imaging (rs-fMRI) a common strategy to reduce the impact of physiological noise and other artifacts on the data is to regress out the global signal using global signal regression (GSR). Yet, GSR is one of the most controversial preprocessing techniques for rs-fMRI. It effectively removes non-neuronal artifacts, but at the same time it alters correlational patterns in unpredicted ways. Furthermore the global signal includes neural BOLD signal by construction, and is consequently related to neural and behavioral function. Performing GSR taking into account the underlying physiology (mainly the blood arrival time) has been proved to be beneficial. From these observations we aimed to: 1) characterize the effect of GSR on network-level functional connectivity in a large dataset; 2) assess the complementary role of global signal and vessels; and 3) use the framework of partial information decomposition to further look into the joint dynamics of the global signal and vessels, and their respective influence on the dynamics of cortical areas. We observe that GSR affects intrinsic connectivity networks in the connectome in a non-uniform way. Furthermore, by estimating the predictive information of blood flow and the global signal using partial information decomposition, we observe that both signals are present in different amounts across intrinsic connectivity networks. Simulations showed that differences in blood arrival time can largely explain this phenomenon. With these results we confirm network-specific effects of GSR and the importance of taking blood flow into account for improve denoising methods. Using GSR but not correcting for blood flow might selectively introduce physiological artifacts across intrinsic connectivity networks that distort the functional connectivity estimates.

neuroscience↗

bioRxiv2019

Static and Dynamic Functional Connectivity Analysis of Cerebrovascular Reactivity: An fMRI Study

The human brain, as a finely-tuned system, needs a constant flow of oxygen to function properly. To accomplish this, the cerebrovascular system ensures a steady stream of oxygenation to brain cells. One tool that the cerebrovascular system uses is cerebrovascular reactivity (CVR), which is the systems ability to react to vasoactive stimuli. Understanding CVR can provide unique information about cerebrovascular diseases and general brain function. CVR can be evaluated by scanning subjects with blood-oxygenation-level-dependent (BOLD) functional magnetic resonance imaging (fMRI) while they periodically inhale room air and CO2-enriched gas, a powerful and widely-used vasodilator. Our goal is to understand the effect of vasodilation on individual intrinsic connectivity networks (ICNs), as well as how functional network connectivity (FNC) adapts to the same vasodilation. To achieve this goal, we first developed an innovative metric to measure the effect of CVR on ICNs, which contrasts to the commonly used voxel-wise CVR. Furthermore, for the first time, we studied static (sFNC) and dynamic (dFNC) FNC in the context of CVR. Our results show that network connectivity is generally weaker during vascular dilation, and these results are more pronounced in dFNC analysis. dFNC analysis reveals that participants did not return to the pre-CO2 inhalation state, suggesting that the one-minute period of room-air inhalation is not enough for the CO2 effect to fully dissipate in humans. Overall, we see new relationships between CVR and ICNs, as well as how FNC adapts to vascular system changes.

neuroscience↗

bioRxiv2019

Unraveling Diagnostic Biomarkers of Schizophrenia through Structure-Revealing Fusion of Multi-Modal Neuroimaging Data

Fusing complementary information from different modalities can lead to the discovery of more accurate diagnostic biomarkers for psychiatric disorders. However, biomarker discovery through data fusion is challenging since it requires extracting interpretable and reproducible patterns from data sets, consisting of shared/unshared patterns and of different orders. For example, multi-channel electroencephalography (EEG) signals from multiple subjects can be represented as a third-order tensor with modes: subject, time, and channel, while functional magnetic resonance imaging (fMRI) data may be in the form of subject by voxel matrices. Traditional data fusion methods rearrange higher-order tensors, such as EEG, as matrices to use matrix factorization-based approaches. In contrast, fusion methods based on coupled matrix and tensor factorizations (CMTF) exploit the potential multi-way structure of higher-order tensors. The CMTF approach has been shown to capture underlying patterns more accurately without imposing strong constraints on the latent neural patterns, i.e., biomarkers. In this paper, EEG, fMRI and structural MRI (sMRI) data collected during an auditory oddball task (AOD) from a group of subjects consisting of patients with schizophrenia and healthy controls, are arranged as matrices and higher-order tensors coupled along the subject mode, and jointly analyzed using structure-revealing CMTF methods (also known as advanced CMTF (ACMTF)) focusing on unique identification of underlying patterns in the presence of shared/unshared patterns. We demonstrate that joint analysis of the EEG tensor and fMRI matrix using ACMTF reveals significant and biologically meaningful components in terms of differentiating between patients with schizophrenia and healthy controls while also providing spatial patterns with high resolution and improving the clustering performance compared to the analysis of only the EEG tensor. We also show that these patterns are reproducible, and study reproducibility for different model parameters. In comparison to the joint independent component analysis (jICA) data fusion approach, ACMTF provides easier interpretation of EEG data by revealing a single summary map of the topography for each component. Furthermore, fusion of sMRI data with EEG and fMRI through an ACMTF model provides structural patterns; however, we also show that when fusing data sets from multiple modalities, hence of very different nature, preprocessing plays a crucial role.

neuroscience↗