Dynamic Functional Connectivity Research Articles

ID: 331943 Subcortical to Neocortical Dynamic Functional Connectivity Differs in Patients with Focal Epilepsy: Opportunities for Neuromodulation

ID: 331943 Subcortical to Neocortical Dynamic Functional Connectivity Differs in Patients with Focal Epilepsy: Opportunities for Neuromodulation

Abnormal thalamocortical network dynamics in patients with migraine and its relationship with electroacupuncture treatment response.

Acupuncture is an effective and safe alternative treatment to prevent and treat migraine, but its central analgesic mechanism remains poorly understood. It is believed that the dysfunction of the thalamocortical connectivity network is an important contributor to migraine pathophysiology. This study aimed to investigate the abnormal thalamocortical network dynamics in patients with migraine without aura (MWoA) before and after an 8-week electroacupuncture treatment. A total of 143 patients with MWoA and 100 healthy controls (HC) were included, and resting-state functional magnetic resonance imaging (fMRI) data were acquired. Dynamic functional network connectivity (dFNC) was calculated for each subject. The modulation effect of electroacupuncture on clinical outcomes of migraine, dFNC, and their association were investigated. In our results, dFNC matrices were classified into two clusters (brain states). As compared with the HC, patients with MWoA had a higher proportion of brain states with a strong thalamocortical between-network connection, implying an abnormal balance of the network organization across dFNC brain states. Correlation analysis showed that this abnormality was associated with summarized clinical measurements of migraine. A total of 60 patients were willing to receive an 8-week electroacupuncture treatment, and 24 responders had 50% changes in headache frequency. In electroacupuncture responders, electroacupuncture could change the abnormal thalamocortical connectivities towards a pattern more similar to that of HC. Our findings suggested that electroacupuncture could relieve the symptoms of migraine and has the potential capacity to regulate the abnormal function of the thalamocortical circuits.

Characterization of the temporal stability of ToM and pain functional brain networks carry distinct developmental signatures during naturalistic viewing

A temporally stable functional brain network pattern among coordinated brain regions is fundamental to stimulus selectivity and functional specificity during the critical period of brain development. Brain networks that are recruited in time to process internal states of others’ bodies (like hunger and pain) versus internal mental states (like beliefs, desires, and emotions) of others’ minds allow us to ask whether a quantitative characterization of the stability of these networks carries meaning during early development and constrain cognition in a specific way. Previous research provides critical insight into the early development of the theory-of-mind (ToM) network and its segregation from the Pain network throughout normal development using functional connectivity. However, a quantitative characterization of the temporal stability of ToM networks from early childhood to adulthood remains unexplored. In this work, reusing a large sample of children (n = 122, 3–12 years) and adults (n = 33) dataset that is available on the OpenfMRI database under the accession number ds000228, we addressed this question based on their fMRI data during a short and engaging naturalistic movie-watching task. The movie highlights the characters’ bodily sensations (often pain) and mental states (beliefs, desires, emotions), and is a feasible experiment for young children. Our results tracked the change in temporal stability using an unsupervised characterization of ToM and Pain networks DFC patterns using Angular and Mahalanobis distances between dominant dynamic functional connectivity subspaces. Our findings reveal that both ToM and Pain networks exhibit lower temporal stability as early as 3-years and gradually stabilize by 5-years, which continues till adolescence and late adulthood (often sharing similarity with adult DFC stability patterns). Furthermore, we find that the temporal stability of ToM brain networks is associated with the performance of participants in the false belief task to access mentalization at an early age. Interestingly, higher temporal stability is associated with the pass group, and similarly, moderate and low temporal stability are associated with the inconsistent group and the fail group. Our findings open an avenue for applying the temporal stability of large-scale functional brain networks during cortical development to act as a biomarker for multiple developmental disorders concerning impairment and discontinuity in the neural basis of social cognition.

RS-MAE: Region-State Masked Autoencoder for Neuropsychiatric Disorder Classifications Based on Resting-State fMRI.

Dynamic functional connectivity (DFC) extracted from resting-state functional magnetic resonance imaging (fMRI) has been widely used for neuropsychiatric disorder classifications. However, serious information redundancy within DFC matrices can significantly undermine the performance of classification models based on them. Moreover, traditional deep models cannot adapt well to connectivity-like data, and insufficient training samples further hinder their effective training. In this study, we proposed a novel region-state masked autoencoder (RS-MAE) for proficient representation learning based on DFC matrices and ultimately neuropsychiatric disorder classifications based on fMRI. Three strategies were taken to address the aforementioned limitations. First, masked autoencoder (MAE) was introduced to reduce redundancy within DFC matrices and learn effective representations of human brain function simultaneously. Second, region-state (RS) patch embedding was proposed to replace space-time patch embedding in video MAE to adapt to DFC matrices, in which only topological locality, rather than spatial locality, exists. Third, random state concatenation (RSC) was introduced as a DFC matrix augmentation approach, to alleviate the problem of training sample insufficiency. Neuropsychiatric disorder classifications were attained by fine-tuning the pretrained encoder included in RS-MAE. The performance of the proposed RS-MAE was evaluated on four publicly available datasets, achieving accuracies of 76.32%, 77.25%, 88.87%, and 76.53% for the attention deficit and hyperactivity disorder (ADHD), autism spectrum disorder (ASD), Alzheimer's disease (AD), and schizophrenia (SCZ) classification tasks, respectively. These results demonstrate the efficacy of the RS-MAE as a proficient deep learning model for neuropsychiatric disorder classifications.

Dynamic functional connectivity in verbal cognitive control and word reading

Cognitive control processes enable the suppression of automatic behaviors and the initiation of appropriate responses. The Stroop color naming task serves as a benchmark paradigm for understanding the neurobiological model of verbal cognitive control. Previous research indicates a predominant engagement of the prefrontal and premotor cortex during the Stroop task compared to reading. We aim to further this understanding by creating a dynamic atlas of task-preferential modulations of functional connectivity through white matter. Patients undertook word-reading and Stroop tasks during intracranial EEG recording. We quantified task-related high-gamma amplitude modulations at 547 nonepileptic electrode sites, and a mixed model analysis identified regions and timeframes where these amplitudes differed between tasks. We then visualized white matter pathways with task-preferential functional connectivity enhancements at given moments. Word reading, compared to the Stroop task, exhibited enhanced functional connectivity in inter- and intra-hemispheric white matter pathways from the left occipital-temporal region 350–600 ms before response, including the posterior callosal fibers as well as the left vertical occipital, inferior longitudinal, inferior fronto-occipital, and arcuate fasciculi. The Stroop task showed enhanced functional connectivity in the pathways from the left middle-frontal pre-central gyri, involving the left frontal u-fibers and anterior callosal fibers. Automatic word reading largely utilizes the left occipital-temporal cortices and associated white matter tracts. Verbal cognitive control predominantly involves the left middle frontal and precentral gyri and its connected pathways. Our dynamic tractography atlases may serve as a novel resource providing insights into the unique neural dynamics and pathways of automatic reading and verbal cognitive control.

Dynamic functional connections analysis with spectral learning for brain disorder detection

Dynamic functional connections (dFCs), can reveal neural activities, which provides an insightful way of mining the temporal patterns within the human brain and further detecting brain disorders. However, most existing studies focus on the dFCs estimation to identify brain disorders by shallow temporal features and methods, which cannot capture the inherent temporal patterns of dFCs effectively. To address this problem, this study proposes a novel method, named dynamic functional connections analysis with spectral learning (dCSL), to explore inherently temporal patterns of dFCs and further detect the brain disorders. Concretely, dCSL includes two components, dFCs estimation module and dFCs analysis module. In the former, dFCs are estimated via the sliding window technique. In the latter, the spectral kernel mapping is first constructed by combining the Fourier transform with the non-stationary kernel. Subsequently, the spectral kernel mapping is stacked into a deep kernel network to explore higher-order temporal patterns of dFCs through spectral learning. The proposed dCSL, sharing the benefits of deep architecture and non-stationary kernel, can not only calculate the long-range relationship but also explore the higher-order temporal patterns of dFCs. To evaluate the proposed method, a set of brain disorder classification tasks are conducted on several public datasets. As a result, the proposed dCSL achieves 5% accuracy improvement compared with the widely used approaches for analyzing sequence data, 1.3% accuracy improvement compared with the state-of-the-art methods for dFCs. In addition, the discriminative brain regions are explored in the ASD detection task. The findings in this study are consistent with the clinical performance in ASD.

The dynamic reconfiguration of the functional network during episodic memory task predicts the memory performance

Episodic memory is essential for forming and retaining personal experiences, representing a fundamental aspect of human cognition. Traditional studies of episodic memory have typically used static analysis methods, viewing the brain as an unchanging entity and overlooking its dynamic properties over time. In this study, we utilized dynamic functional connectivity analysis on fMRI data from healthy adults performing an episodic memory task. We quantified integration and recruitment metrics and examined their correlation with memory performance using Pearson correlation. During encoding, integration across the entire brain, especially within the frontoparietal subnetwork, was significantly correlated with memory performance. During retrieval, recruitment becomes significantly associated with memory performance in visual subnetwork, somatomotor subnetwork, and ventral attention subnetwork. At the nodal level, a significant negative correlation was observed between memory scores and integration of the anterior cingulate gyrus, precentral gyrus, and inferior frontal gyrus within the frontoparietal network during encoding task. During retrieval task, a significant negative correlation was found between memory scores and recruitment in the left progranular cortex and right transverse gyral ventral, whereas positive correlations were seen in the right posterior inferior temporal, left middle temporal, right frontal operculum, and left operculum nodes. Moreover, the dynamic reconfiguration of the functional network was predictive of predict memory performance, as demonstrated by a significant correlation between actual and predicted memory scores. These findings advance our understanding network mechanisms underlying memory processes and developing intervention approaches for memory-related disorders as they shed light on critical factors involved in cognitive processes and provide a deeper understanding of the underlying mechanisms driving cognitive function.

Stable individual differences from dynamic patterns of function: brain network flexibility predicts openness/intellect, intelligence, and psychoticism.

A growing understanding of the nature of brain function has led to increased interest in interpreting the properties of large-scale brain networks. Methodological advances in network neuroscience provide means to decompose these networks into smaller functional communities and measure how they reconfigure over time as an index of their dynamic and flexible properties. Recent evidence has identified associations between flexibility and a variety of traits pertaining to complex cognition including creativity and working memory. The present study used measures of dynamic resting-state functional connectivity in data from the Human Connectome Project (n = 994) to test associations with Openness/Intellect, general intelligence, and psychoticism, three traits that involve flexible cognition. Using a machine-learning cross-validation approach, we identified reliable associations of intelligence with cohesive flexibility of parcels in large communities across the cortex, of psychoticism with disjoint flexibility, and of Openness/Intellect with overall flexibility among parcels in smaller communities. These findings are reasonably consistent with previous theories of the neural correlates of these traits and help to expand on previous associations of behavior with dynamic functional connectivity, in the context of broad personality dimensions.

Dynamic functional network connectivity in children with profound bilateral congenital sensorineural hearing loss.

Functional magnetic resonance imaging (fMRI) studies have revealed extensive functional reorganization in patients with sensorineural hearing loss (SNHL). However, almost no study focuses on the dynamic functional connectivity after hearing loss. This study aimed to investigate dynamic functional connectivity changes in children with profound bilateral congenital SNHL under the age of 3years. Thirty-two children with profound bilateral congenital SNHL and 24 children with normal hearing were recruited for the present study. Independent component analysis identified 18 independent components composing five resting-state networks. A sliding window approach was used to acquire dynamic functional matrices. Three states were identified using the k-means algorithm. Then, the differences in temporal properties and the variance of network efficiency between groups were compared. The children with SNHL showed longer mean dwell time and decreased functional connectivity between the auditory network and sensorimotor network in state 3 (P < 0.05), which was characterized by relatively stronger functional connectivity between high-order resting-state networks and motion and perception networks. There was no difference in the variance of network efficiency. These results indicated the functional reorganization due to hearing loss. This study also provided new perspectives for understanding the state-dependent connectivity patterns in children with SNHL.

Dynamic functional network connectivity and its association with lipid metabolism in Alzheimer's disease.

The study aims to examine the changing trajectory characteristics of dynamic functional network connectivity (dFNC) and its correlation with lipid metabolism-related factors across the Alzheimer's disease (AD) spectrum populations. Data from 242 AD spectrum subjects, including biological, neuroimaging, and general cognition, were obtained from the Alzheimer's Disease Neuroimaging Initiative for this cross-sectional study. The study utilized a sliding-window approach to assess whole-brain dFNC, investigating group differences and associations with biological and cognitive factors. Abnormal dFNC was used in the classification of AD spectrum populations by support vector machine. Mediation analysis was performed to explore the relationships between lipid-related indicators, dFNC, cerebrospinal fluid (CSF) biomarkers, and cognitive performance. Significant group difference concerning were observed in relation to APOE-ε4 status, CSF biomarkers, and cognitive scores. Two reoccurring connectivity states were identified: state-1 characterized by frequent but weak connections, and state-II characterized by less frequent but strong connections. Pre-AD subjects exhibited a preference for spending more time in state-I, whereas AD patients tended remain in state-II for longer periods. Group difference in dFNC was primarily found between AD and non-AD participants within each state. The dFNC of state-I yielded strong power to distinguish AD from other groups compared with state-II. APOE-ε4+, high polygenic score, and high serum lipid group were strongly associated with network disruption between association cortex system and sensory cortex system that characterized elevation of cognitive function, which may suggest a compensatory mechanism of dFNC in state-I, whereas differential connections of state-II mediated the relationships between APOE-ε4 genotype and CSF biomarkers, and cognitive indicators. The dysfunction of dFNC temporal-spatial patterns and increased cognition in individuals with APOE-ε4, high polygenic score, and higher serum lipid levels shed light on the lipid-related mechanisms of dynamic network reorganization in AD.

Dynamic Functional Network Connectivity in Acute Incomplete Cervical Cord Injury Patients and Its Associations With Sensorimotor Dysfunction Measures

BackgroundDynamic functional network connectivity (dFNC) captures temporal variations in functional connectivity during MRI acquisition. However, the neural mechanisms driving dFNC alterations in the brain networks of patients with Acute incomplete cervical cord injury (AICCI) remain unclear. MethodsThis study included 16 AICCI patients and 16 healthy controls (HC). Initially, Independent Component Analysis (ICA) was employed to extract whole-brain independent components (ICs) from resting-state functional MRI (rs-fMRI) data. Subsequently, a sliding time window approach, combined with k-means clustering, was used to estimate dFNC states for each participant. Finally, a correlation analysis was conducted to examine the association between sensorimotor dysfunction scores in AICCI patients and the temporal characteristics of dFNC. ResultICA was employed to extract 26 whole-brain ICs. Subsequent dynamic analysis identified four distinct connectivity states across the entire cohort. Notably, AICCI patients demonstrated a significant preference for State 3 compared to HC, as evidenced by a higher frequency and longer duration spent in this state. Conversely, State 4 exhibited a reduced frequency and shorter dwell time in AICCI patients. Moreover, correlation analysis revealed a positive association between sensorimotor dysfunction and both the mean dwell time and the fractional of time spent in State 3. ConclusionsPatients with AICCI demonstrate abnormal connectivity within dFNC states, and the temporal characteristics of dFNC are associated with sensorimotor dysfunction scores. These findings highlight the potential of dFNC as a sensitive biomarker for detecting network functional changes in AICCI patients, providing valuable insights into the dynamic alterations in brain connectivity related to sensorimotor dysfunction in this population.

Multi-Domain Based Dynamic Graph Representation Learning for EEG Emotion Recognition.

Graph neural networks (GNNs) have demonstrated efficient processing of graph-structured data, making them a promising method for electroencephalogram (EEG) emotion recognition. However, due to dynamic functional connectivity and nonlinear relationships between brain regions, representing EEG as graph data remains a great challenge. To solve this problem, we proposed a multi-domain based graph representation learning (MD 2GRL) framework to model EEG signals as graph data. Specifically, MD 2GRL leverages gated recurrent units (GRU) and power spectral density (PSD) to construct node features of two subgraphs. Subsequently, the self-attention mechanism is adopted to learn the similarity matrix between nodes and fuse it with the intrinsic spatial matrix of EEG to compute the corresponding adjacency matrix. In addition, we introduced a learnable soft thresholding operator to sparsify the adjacency matrix to reduce noise in the graph structure. In the downstream task, we designed a dual-branch GNN and incorporated spatial asymmetry for graph coarsening. We conducted experiments using the publicly available datasets SEED and DEAP, separately for subject-dependent and subject-independent, to evaluate the performance of our model in emotion classification. Experimental results demonstrated that our method achieved state-of-the-art (SOTA) classification performance in both subject-dependent and subject-independent experiments. Furthermore, the visualization analysis of the learned graph structure reveals EEG channel connections that are significantly related to emotion and suppress irrelevant noise. These findings are consistent with established neuroscience research and demonstrate the potential of our approach in comprehending the neural underpinnings of emotion.

A Plug-In Graph Neural Network to Boost Temporal Sensitivity in fMRI Analysis.

Learning-based methods offer performance leaps over traditional methods in classification analysis of high-dimensional functional MRI (fMRI) data. In this domain, deep-learning models that analyze functional connectivity (FC) features among brain regions have been particularly promising. However, many existing models receive as input temporally static FC features that summarize inter-regional interactions across an entire scan, reducing the temporal sensitivity of classifiers by limiting their ability to leverage information on dynamic FC features of brain activity. To improve the performance of baseline classification models without compromising efficiency, here we propose a novel plug-in based on a graph neural network, GraphCorr, to provide enhanced input features to baseline models. The proposed plug-in computes a set of latent FC features with enhanced temporal information while maintaining comparable dimensionality to static features. Taking brain regions as nodes and blood-oxygen-level-dependent (BOLD) signals as node inputs, GraphCorr leverages a node embedder module based on a transformer encoder to capture dynamic latent representations of BOLD signals. GraphCorr also leverages a lag filter module to account for delayed interactions across nodes by learning correlational features of windowed BOLD signals across time delays. These two feature groups are then fused via a message passing algorithm executed on the formulated graph. Comprehensive demonstrations on three public datasets indicate improved classification performance for several state-of-the-art graph and convolutional baseline models when they are augmented with GraphCorr.

Static and Dynamic Dysconnectivity in Early Psychosis: Relationship With Symptom Dimensions.

Altered functional connectivity (FC) has been frequently reported in psychosis. Studying FC and its time-varying patterns in early-stage psychosis allows the investigation of the neural mechanisms of this disorder without the confounding effects of drug treatment or illness-related factors. We employed resting-state functional magnetic resonance imaging (rs-fMRI) to explore FC in individuals with early psychosis (EP), who also underwent clinical and neuropsychological assessments. 96 EP and 56 demographically matched healthy controls (HC) from the Human Connectome Project for Early Psychosis database were included. Multivariate analyses using spatial group independent component analysis were used to compute static FC and dynamic functional network connectivity (dFNC). Partial correlations between FC measures and clinical and cognitive variables were performed to test brain-behavior associations. Compared to HC, EP showed higher static FC in the striatum and temporal, frontal, and parietal cortex, as well as lower FC in the frontal, parietal, and occipital gyrus. We found a negative correlation in EP between cognitive function and FC in the right striatum FC (pFWE = 0.009). All dFNC parameters, including dynamism and fluidity measures, were altered in EP, and positive symptoms were negatively correlated with the meta-state changes and the total distance (pFWE = 0.040 and pFWE = 0.049). Our findings support the view that psychosis is characterized from the early stages by complex alterations in intrinsic static and dynamic FC, that may ultimately result in positive symptoms and cognitive deficits.

Altered functional connectivity and spatiotemporal dynamics in individuals with sleep disorders.

Idiopathic hypersomnia (IH) is a sleep disorder characterized by highly disruptive symptoms. Like narcolepsy type 1, a well-characterized sleep disorder, individuals with IH suffer from excessive daytime sleepiness, though there is little overlap in metabolic or neural biomarkers across these two disorders. This lack of common pathophysiology, combined with the clear overlap in symptoms presents an ideal paradigm for better understanding the impact of IH on an individual's functional activity and organization, and potentially, the underlying pathophysiology. This study examines the observed functional connectivity in patients with IH, and patients with narcolepsy type 1 (NT1) against healthy control individuals. Static functional connectivity is compared, as are quasi-periodic patterns, acquired from the BOLD timecourse, for all groups. In addition to baseline data comparison, the study also included a post-nap condition, where the individuals included in this analysis napped for at least 10 minutes prior to the scanning session, to explore why individuals with IH do not feel refreshed after a nap like individuals with NT1 do. Assessing the groups' spatiotemporal patterns revealed key differences across both disorders and conditions: static connectivity revealed at baseline higher subcortical connectivity in the NT1 group. There was also observably less connectivity in the IH group both at baseline and post-nap, though none of these static analyses survived multiple comparisons correction to reach significance. The QPP results however found significant differences in the IH group in key networks, particularly the DAN/FPCN correlation is significantly different at baseline vs. post-nap, a trend not observed in either the control or NT1 groups. The DAN and FPCN are drastically altered both at baseline and post-nap when compared to the other groups, and may likely be a disorder-specific result. This study demonstrates that key networks for arousal are more heavily disrupted in IH patients, who are less affected by a nap, confirmed through both subject reporting and functional evidence through spatiotemporal patterns.

Inflammation-related abnormal dynamic brain activity correlates with cognitive impairment in first-episode, drug-naïve major depressive disorder

BackgroundCognitive impairment is common in major depressive disorder (MDD) and potentially linked to inflammation-induced alterations in brain function. However, the relationship between inflammation, dynamic brain activity, and cognitive impairment in MDD remains unclear. MethodsFifty-seven first-episode, drug-naïve MDD patients and sixty healthy controls underwent fMRI scanning. Dynamic amplitude of low-frequency fluctuations (dALFF) and dynamic functional connectivity (dFC) were measured using the sliding window method. Plasma IL − 6 levels and cognitive function were assessed using enzyme-linked immunosorbent assay (ELISA) and the Repeated Battery for Assessment of Neuropsychological Status (RBANS), respectively. ResultsMDD patients exhibited decreased dALFF in the bilateral inferior temporal gyrus (ITG), right inferior frontal gyrus, opercular part (IFGoperc), and bilateral middle occipital gyrus (MOG). Regions of dALFF associated with IL-6 included right ITG (r = −0.400/p = 0.003), left ITG (r = −0.381/p = 0.004), right IFGoperc (r = −0.342/p = 0.011), and right MOG (r = −0.327/p = 0.016). Furthermore, IL-6-related abnormal dALFF (including right ITG: r = 0.309/p = 0.023, left ITG: r = 0.276/p = 0.044) was associated with attention impairment. These associations were absent entirely in MDD patients without suicidal ideation. Additionally, IL-6 levels were correlated with dFC of specific brain regions. LimitationsSmall sample size and cross-sectional study design. ConclusionsInflammation-related dALFF was associated with attention impairment in MDD patients, with variations observed among MDD subgroups. These findings contribute to the understanding of the intricate relationship between inflammation, dynamic brain activity and cognitive impairments in MDD.

Repetitive transcranial magnetic stimulation reshaped the dynamic reconfiguration of the executive and reward networks in individuals with tobacco use disorder

BackgroundStudies have demonstrated the potential of repetitive transcranial magnetic stimulation (rTMS) to decrease smoking cravings in individuals with tobacco use disorder (TUD). However, the neural features underlying the effects of rTMS treatment, especially the dynamic attributes of brain networks associated with the treatment, remain unclear. MethodsUsing dynamic functional connectivity analysis, this study first explored the differences in dynamic functional network features between 60 subjects with TUD and 64 nonsmoking healthy controls (HCs). Then, the left dorsolateral prefrontal cortex (DLPFC) was targeted for a five-day course of rTMS treatment in the 60 subjects with TUD (active rTMS in 42 subjects and sham treatment in 18 subjects). We explored the effect of rTMS on the dynamic network features associated with rTMS by comparing the actively treated group and the sham group. ResultsCompared to nonsmokers, TUD subjects exhibited an increased integration coefficient between the frontoparietal network (FPN) and the basal ganglia network (BGN) and a reduced integration coefficient between the medial frontal network (MFN) and the FPN. Analysis of variance revealed that rTMS treatment reduced the integration coefficient between the FPN and BGN and improved the recruitment coefficient of the FPN. LimitationsThis study involved a limited sample of young male smokers, and the findings may not generalize to older smokers or female smokers with an extensive history of smoking. ConclusionrTMS treatment of the left DLPFC exhibited significant effectiveness in restructuring the neural circuits associated with TUD while significantly mitigating smoking cravings.

Altered cerebellar-cerebral dynamic functional connectivity in patients with pontine stroke: a resting-state fMRI study.

Potential changes in patterns of dynamic functional network connections at the cerebellar-cerebral level in pontine infarction (PI) patients remain unclear. The study aimed to investigate the abnormal patterns of dynamic functional connectivity (dFC) between the cerebellar subregions within networks and regions of the cerebral cortex in patients with PI. Forty-six chronic left pontine infarction (LPI), 32 chronic right pontine infarction (RPI), and 50 healthy controls (HCs) were recruited to undergo resting-state fMRI scans. Cerebellar-cerebral dFC was characterized using the sliding window method and seed-based connectivity analyses. Correlations between altered dFC values and clinical variables (The Rey Auditory Verbal Learning Test and Flanker task) in PI patients and healthy controls were investigated. Compared with HCs, the PI groups showed significantly aberrant cerebellar-cerebral dFC between cerebellar subregions within networks and supratentorial cerebral cortex, including executive, default-mode, and motor networks. Furthermore, Correlation analysis showed a decoupling between abnormal dFC and cognitive functions in PI patients. These findings indicate that PI patients are accompanied by damage to cerebellar subregions within networks and cerebellar-cerebral pathways, which may provide a potential target for treatment or an indication of therapeutic efficacy.

Altered dynamic functional connectivity of nucleus accumbens subregions in major depressive disorder: the interactive effect of childhood trauma and diagnosis.

Major depressive disorder (MDD) with childhood trauma represents a heterogeneous clinical subtype of depression. Previous research has observed alterations in the reward circuitry centered around the nucleus accumbens (NAc) in MDD patients. However, limited investigations have focused on aberrant functional connectivity (FC) within NAc subregions among MDD with childhood trauma. Thus, this study adopts analyses of both static FC (sFC) and dynamic FC (dFC) to examine neurobiological changes in MDD with childhood trauma. The bilateral nucleus accumbens shell (NAc-shell) and nucleus accumbens core (NAc-core) were selected as the seeds. Four participant groups were included: MDD with childhood trauma (n = 48), MDD without childhood trauma (n = 30), healthy controls (HCs) with childhood trauma (n = 57), and HCs without childhood trauma (n = 46). Our findings revealed both abnormal sFC and dFC between NAc-shell and NAc-core and regions including the middle occipital gyrus (MOG), anterior cingulate cortex, and inferior frontal gyrus in MDD with childhood trauma. Furthermore, a significant correlation was identified between the dFC of the left NAc-shell and the right MOG in relation to childhood trauma. Additionally, abnormal dFC moderated the link between childhood abuse and depression severity. These outcomes shed light on the neurobiological underpinnings of MDD with childhood trauma.

Stacking models of brain dynamics to improve prediction of subject traits in fMRI

Abstract Beyond structural and time-averaged functional connectivity brain measures, modelling the way brain activity dynamically unfolds can add important information to our understanding and characterisation of individual cognitive traits. One approach to leveraging this information is to extract features from models of brain network dynamics to predict individual traits. However, these predictions are susceptible to variability due to factors such as variation in model estimation induced by the choice of hyperparameters. We suggest that, rather than merely being statistical noise, this variability may be useful in providing complementary information that can be leveraged to improve prediction accuracy. To leverage this variability, we propose the use of stacking, a prediction-driven approach for model selection. Specifically, we combine predictions developed from multiple hidden Markov models—a probabilistic generative model of network dynamics that identifies recurring patterns of brain activity—to demonstrate that stacking can slightly improve the accuracy and robustness of cognitive trait predictions. By comparing analysis from the Human Connectome Project and UK Biobank datasets, we show that stacking is relatively effective at improving prediction accuracy and robustness when there are enough subjects, and that the effectiveness of combining predictions from static and dynamic functional connectivity approaches depends on the length of scan per subject. We also show that the effectiveness of stacking predictions is driven by the accuracy and diversity in the underlying model estimations.