Neuroimaging-based Individualized Prediction of Cognition and Behavior for Mental Disorders and Health: Methods and Promises

Neuroimage analysis using artificial intelligence approaches: a systematic review

In the contemporary era, artificial intelligence (AI) has undergone a transformative evolution, exerting a profound influence on neuroimaging data analysis. This development has significantly elevated our comprehension of intricate brain functions. This study investigates the ramifications of employing AI techniques on neuroimaging data, with a specific objective to improve diagnostic capabilities and contribute to the overall progress of the field. A systematic search was conducted in prominent scientific databases, including PubMed, IEEE Xplore, and Scopus, meticulously curating 456 relevant articles on AI-driven neuroimaging analysis spanning from 2013 to 2023. To maintain rigor and credibility, stringent inclusion criteria, quality assessments, and precise data extraction protocols were consistently enforced throughout this review. Following a rigorous selection process, 104 studies were selected for review, focusing on diverse neuroimaging modalities with an emphasis on mental and neurological disorders. Among these, 19.2% addressed mental illness, and 80.7% focused on neurological disorders. It is found that the prevailing clinical tasks are disease classification (58.7%) and lesion segmentation (28.9%), whereas image reconstruction constituted 7.3%, and image regression and prediction tasks represented 9.6%. AI-driven neuroimaging analysis holds tremendous potential, transforming both research and clinical applications. Machine learning and deep learning algorithms outperform traditional methods, reshaping the field significantly.

Multimodal frontal neuroimaging markers predict longitudinal craving reduction in abstinent individuals with heroin use disorder

The variation in improvement among individuals with addiction after abstinence is a critical issue. Here, we aimed to identify robust multimodal markers associated with high response to 8-month abstinence in the individuals with heroin use disorder (HUD) and explore whether the identified markers could be generalized to the individuals with methamphetamine use disorder (MUD). According to the median of craving changes, 53 individuals with HUD with 8-month abstinence were divided into two groups: higher craving reduction and lower craving reduction. At baseline, clinical variables, cortical thickness and subcortical volume, fractional anisotropy (FA) of fibers and resting-state functional connectivity (RSFC) were extracted. Different strategies (single metric, multimodal neuroimaging fusion and multimodal neuroimaging-clinical data fusion) were used to identify reliable features for discriminating the individuals with HUD with higher craving reduction from those with lower reduction. The generalization ability of the identified features was validated in the 21 individuals with MUD. Multimodal neuroimaging-clinical fusion features with best performance was achieved an 87.1 ± 3.89% average accuracy in individuals with HUD, with a moderate accuracy of 66.7% when generalizing to individuals with MUD. The multimodal neuroimaging features, primarily converging in frontal regions (e.g., the left superior frontal (LSF) thickness, FA of the LSF-occipital tract, and RSFC of left middle frontal-right superior temporal lobe), collectively contributed to prediction alongside dosage and attention impulsiveness. In this study, we identified the validated multimodal frontal neuroimaging markers associated with higher response to long-term abstinence and revealed insights for the neural mechanisms of addiction abstinence, contributing to clinical strategies and treatment for addiction.

Replicability and generalizability in population psychiatric neuroimaging

Studies linking mental health with brain function in cross-sectional population-based association studies have historically relied on small, underpowered samples. Given the small effect sizes typical of such brain-wide associations, studies require samples into the thousands to achieve the statistical power necessary for replicability. Here, we detail how small sample sizes have hampered replicability and provide sample size targets given established association strength benchmarks. Critically, while replicability will improve with larger samples, it is not guaranteed that observed effects will meaningfully apply to target populations of interest (i.e., be generalizable). We discuss important considerations related to generalizability in psychiatric neuroimaging and provide an example of generalizability failure due to "shortcut learning" in brain-based predictions of mental health phenotypes. Shortcut learning is a phenomenon whereby machine learning models learn an association between the brain and an unmeasured construct (the shortcut), rather than the intended target of mental health. Given the complex nature of brain-behavior interactions, the future of epidemiological approaches to brain-based studies of mental health will require large, diverse samples with comprehensive assessment.

Gray matters: ViT-GAN framework for identifying schizophrenia biomarkers linking structural MRI and functional network connectivity

Brain disorders are often associated with changes in brain structure and function, where functional changes may be due to underlying structural variations. Gray matter (GM) volume segmentation from 3D structural MRI offers vital structural information for brain disorders like schizophrenia, as it encompasses essential brain tissues such as neuronal cell bodies, dendrites, and synapses, which are crucial for neural signal processing and transmission; changes in GM volume can thus indicate alterations in these tissues, reflecting underlying pathological conditions. In addition, the use of the ICA algorithm to transform high-dimensional fMRI data into functional network connectivity (FNC) matrices serves as an effective carrier of functional information. In our study, we introduce a new generative deep learning architecture, the conditional efficient vision transformer generative adversarial network (cEViT-GAN), which adeptly generates FNC matrices conditioned on GM to facilitate the exploration of potential connections between brain structure and function. We developed a new, lightweight self-attention mechanism for our ViT-based generator, enhancing the generation of refined attention maps critical for identifying structural biomarkers based on GM. Our approach not only generates high quality FNC matrices with a Pearson correlation of 0.74 compared to real FNC data, but also uses attention map technology to identify potential biomarkers in GM structure that could lead to functional abnormalities in schizophrenia patients. Visualization experiments within our study have highlighted these structural biomarkers, including the medial prefrontal cortex (mPFC), dorsolateral prefrontal cortex (DL-PFC), and cerebellum. In addition, through cross-domain analysis comparing generated and real FNC matrices, we have identified functional connections with the highest correlations to structural information, further validating the structure-function connections. This comprehensive analysis helps to understand the intricate relationship between brain structure and its functional manifestations, providing a more refined insight into the neurobiological research of schizophrenia.

Social Cognition and Functional Connectivity in Early and Chronic Schizophrenia

BACKGROUND

Individuals with schizophrenia (SZ) experience impairments in social cognition that contribute to poor functional outcomes. However, mechanisms of social cognitive dysfunction in SZ remain poorly understood, which impedes the design of novel interventions to improve outcomes. This pre-registered project (https://doi.org/10.17605/OSF.IO/JH5FC) examines the representation of social cognition in the brain's functional architecture across early and chronic SZ.

METHODS

The study contains two parts: a confirmatory and an exploratory portion. In the confirmatory portion, we identified resting-state connectivity disruptions evident in early and chronic SZ. We performed a connectivity analysis using regions associated with social cognitive dysfunction in early and chronic SZ to test whether aberrant connectivity observed in chronic SZ (N=47; HC=52) was also present in early SZ (N=71, HC=47). In the exploratory portion, we assessed the out-of-sample generalizability and precision of predictive models of social cognition. We used machine learning to predict social cognition and established generalizability with out-of-sample testing and confound control.

RESULTS

Results reveal decreases between left inferior frontal gyrus and intraparietal sulcus in early and chronic SZ, which are significantly associated with social and general cognition and global functioning in chronic SZ and with general cognition and global functioning in early SZ. Predictive modeling reveals the importance of out-of-sample evaluation and confound control.

CONCLUSION

This work provides insights into the functional architecture in early and chronic SZ and suggests that IFG-IPS connectivity could be a prognostic biomarker of social impairments and a target for future interventions (e.g. neuromodulation) focused on improved social functioning.

Morphological Brain Networks of White Matter: Mapping, Evaluation, Characterization, and Application

Although white matter (WM) accounts for nearly half of adult brain, its wiring diagram is largely unknown. Here, an approach is developed to construct WM networks by estimating interregional morphological similarity based on structural magnetic resonance imaging. It is found that morphological WM networks showed nontrivial topology, presented good-to-excellent test-retest reliability, accounted for phenotypic interindividual differences in cognition, and are under genetic control. Through integration with multimodal and multiscale data, it is further showed that morphological WM networks are able to predict the patterns of hamodynamic coherence, metabolic synchronization, gene co-expression, and chemoarchitectonic covariance, and associated with structural connectivity. Moreover, the prediction followed WM functional connectomic hierarchy for the hamodynamic coherence, is related to genes enriched in the forebrain neuron development and differentiation for the gene co-expression, and is associated with serotonergic system-related receptors and transporters for the chemoarchitectonic covariance. Finally, applying this approach to multiple sclerosis and neuromyelitis optica spectrum disorders, it is found that both diseases exhibited morphological dysconnectivity, which are correlated with clinical variables of patients and are able to diagnose and differentiate the diseases. Altogether, these findings indicate that morphological WM networks provide a reliable and biologically meaningful means to explore WM architecture in health and disease.

Machine-learning-based feature selection to identify attention-deficit hyperactivity disorder using whole-brain white matter microstructure: A longitudinal study

BACKGROUND

We aimed to identify important features of white matter microstructures collectively distinguishing individuals with attention-deficit/hyperactivity disorder (ADHD) from those without ADHD using a machine-learning approach.

METHODS

Fifty-one ADHD patients and 60 typically developing controls (TDC) underwent diffusion spectrum imaging at two time points. We evaluated three models to classify ADHD and TDC using various machine-learning algorithms. Model 1 employed baseline white matter features of 45 white matter tracts at Time 1; Model 2 incorporated features from both time points; and Model 3 (main analysis) further included the relative rate of change per year of white matter tracts.

RESULTS

The random forest algorithm demonstrated the best performance for classification. Model 1 achieved an area-under-the-curve (AUC) of 0.67. Model 3, incorporating Time 2 variables and relative rate of change per year, improved the performance (AUC = 0.73). In addition to identifying several white matter features at two time points, we found that the relative rate of change per year in the superior longitudinal fasciculus, frontal aslant tract, stria terminalis, inferior fronto-occipital fasciculus, thalamic and striatal tracts, and other tracts involving sensorimotor regions are important features of ADHD. A higher relative change rate in certain tracts was associated with greater improvement in visual attention, spatial short-term memory, and spatial working memory.

CONCLUSIONS

Our findings support the significant diagnostic value of white matter microstructure and the developmental change rates of specific tracts, reflecting deviations from typical development trajectories, in identifying ADHD.

Multimodal active subspace analysis for computing assessment oriented subspaces from neuroimaging data

BACKGROUND

For successful biomarker discovery, it is essential to develop computational frameworks that summarize high-dimensional neuroimaging data in terms of involved sub-systems of the brain, while also revealing underlying heterogeneous functional and structural changes covarying with specific cognitive and biological traits. However, unsupervised decompositions do not inculcate clinical assessment information, while supervised approaches extract only individual feature importance, thereby impeding qualitative interpretation at the level of subspaces.

NEW METHOD

We present a novel framework to extract robust multimodal brain subspaces associated with changes in a given cognitive or biological trait. Our approach involves active subspace learning on the gradients of a trained machine learning model followed by clustering to extract and summarize the most salient and consistent subspaces associated with the target variable.

RESULTS

Through a rigorous cross-validation procedure on an Alzheimer's disease (AD) dataset, our framework successfully extracts multimodal subspaces specific to a given clinical assessment (e.g., memory and other cognitive skills), and also retains predictive performance in standard machine learning algorithms. We also show that the salient active subspace directions occur consistently across randomly sub-sampled repetitions of the analysis.

COMPARISON WITH EXISTING METHOD(S)

Compared to existing unsupervised decompositions based on principle component analysis, the subspace components in our framework retain higher predictive information.

CONCLUSIONS

As an important step towards biomarker discovery, our framework not only uncovers AD-related brain regions in the associated brain subspaces, but also enables automated identification of multiple underlying structural and functional sub-systems of the brain that collectively characterize changes in memory and proficiency in cognitive skills related to brain disorders like AD.

Never-Ending Learning for Explainable Brain Computing

Exploring the nature of human intelligence and behavior is a longstanding pursuit in cognitive neuroscience, driven by the accumulation of knowledge, information, and data across various studies. However, achieving a unified and transparent interpretation of findings presents formidable challenges. In response, an explainable brain computing framework is proposed that employs the never-ending learning paradigm, integrating evidence combination and fusion computing within a Knowledge-Information-Data (KID) architecture. The framework supports continuous brain cognition investigation, utilizing joint knowledge-driven forward inference and data-driven reverse inference, bolstered by the pre-trained language modeling techniques and the human-in-the-loop mechanisms. In particular, it incorporates internal evidence learning through multi-task functional neuroimaging analyses and external evidence learning via topic modeling of published neuroimaging studies, all of which involve human interactions at different stages. Based on two case studies, the intricate uncertainty surrounding brain localization in human reasoning is revealed. The present study also highlights the potential of systematization to advance explainable brain computing, offering a finer-grained understanding of brain activity patterns related to human intelligence.

Improving Predictability, Test-Retest Reliability and Generalisability of Brain-Wide Associations for Cognitive Abilities via Multimodal Stacking

Brain-wide association studies (BWASs) have attempted to relate cognitive abilities with brain phenotypes, but have been challenged by issues such as predictability, test-retest reliability, and cross-cohort generalisability. To tackle these challenges, we proposed "stacking" that combines brain magnetic resonance imaging of different modalities, from task-fMRI contrasts and functional connectivity during tasks and rest to structural measures, into one prediction model. We benchmarked the benefits of stacking, using the Human Connectome Projects: Young Adults and Aging and the Dunedin Multidisciplinary Health and Development Study. For predictability, stacked models led to out-of-sample r ∼.5-.6 when predicting cognitive abilities at the time of scanning and 36 years earlier. For test-retest reliability, stacked models reached an excellent level of reliability (ICC>.75), even when we stacked only task-fMRI contrasts together. For generalisability, a stacked model with non-task MRI built from one dataset significantly predicted cognitive abilities in other datasets. Altogether, stacking is a viable approach to undertake the three challenges of BWAS for cognitive abilities.

Constructing personalized characterizations of structural brain aberrations in patients with dementia using explainable artificial intelligence

Deep learning approaches for clinical predictions based on magnetic resonance imaging data have shown great promise as a translational technology for diagnosis and prognosis in neurological disorders, but its clinical impact has been limited. This is partially attributed to the opaqueness of deep learning models, causing insufficient understanding of what underlies their decisions. To overcome this, we trained convolutional neural networks on structural brain scans to differentiate dementia patients from healthy controls, and applied layerwise relevance propagation to procure individual-level explanations of the model predictions. Through extensive validations we demonstrate that deviations recognized by the model corroborate existing knowledge of structural brain aberrations in dementia. By employing the explainable dementia classifier in a longitudinal dataset of patients with mild cognitive impairment, we show that the spatially rich explanations complement the model prediction when forecasting transition to dementia and help characterize the biological manifestation of disease in the individual brain. Overall, our work exemplifies the clinical potential of explainable artificial intelligence in precision medicine.

TractGeoNet: A geometric deep learning framework for pointwise analysis of tract microstructure to predict language assessment performance

We propose a geometric deep-learning-based framework, TractGeoNet, for performing regression using diffusion magnetic resonance imaging (dMRI) tractography and associated pointwise tissue microstructure measurements. By employing a point cloud representation, TractGeoNet can directly utilize tissue microstructure and positional information from all points within a fiber tract without the need to average or bin data along the streamline as traditionally required by dMRI tractometry methods. To improve regression performance, we propose a novel loss function, the Paired-Siamese Regression loss, which encourages the model to focus on accurately predicting the relative differences between regression label scores rather than just their absolute values. In addition, to gain insight into the brain regions that contribute most strongly to the prediction results, we propose a Critical Region Localization algorithm. This algorithm identifies highly predictive anatomical regions within the white matter fiber tracts for the regression task. We evaluate the effectiveness of the proposed method by predicting individual performance on two neuropsychological assessments of language using a dataset of 20 association white matter fiber tracts from 806 subjects from the Human Connectome Project Young Adult dataset. The results demonstrate superior prediction performance of TractGeoNet compared to several popular regression models that have been applied to predict individual cognitive performance based on neuroimaging features. Of the twenty tracts studied, we find that the left arcuate fasciculus tract is the most highly predictive of the two studied language performance assessments. Within each tract, we localize critical regions whose microstructure and point information are highly and consistently predictive of language performance across different subjects and across multiple independently trained models. These critical regions are widespread and distributed across both hemispheres and all cerebral lobes, including areas of the brain considered important for language function such as superior and anterior temporal regions, pars opercularis, and precentral gyrus. Overall, TractGeoNet demonstrates the potential of geometric deep learning to enhance the study of the brain's white matter fiber tracts and to relate their structure to human traits such as language performance.

Reliability and predictability of phenotype information from functional connectivity in large imaging datasets

One of the central objectives of contemporary neuroimaging research is to create predictive models that can disentangle the connection between patterns of functional connectivity across the entire brain and various behavioral traits. Previous studies have shown that models trained to predict behavioral features from the individual's functional connectivity have modest to poor performance. In this study, we trained models that predict observable individual traits (phenotypes) and their corresponding singular value decomposition (SVD) representations - herein referred to as latent phenotypes from resting state functional connectivity. For this task, we predicted phenotypes in two large neuroimaging datasets: the Human Connectome Project (HCP) and the Philadelphia Neurodevelopmental Cohort (PNC). We illustrate the importance of regressing out confounds, which could significantly influence phenotype prediction. Our findings reveal that both phenotypes and their corresponding latent phenotypes yield similar predictive performance. Interestingly, only the first five latent phenotypes were reliably identified, and using just these reliable phenotypes for predicting phenotypes yielded a similar performance to using all latent phenotypes. This suggests that the predictable information is present in the first latent phenotypes, allowing the remainder to be filtered out without any harm in performance. This study sheds light on the intricate relationship between functional connectivity and the predictability and reliability of phenotypic information, with potential implications for enhancing predictive modeling in the realm of neuroimaging research.

Triple Interactions Between the Environment, Brain, and Behavior in Children: An ABCD Study

BACKGROUND

Environmental exposures play a crucial role in shaping children's behavioral development. However, the mechanisms by which these exposures interact with brain functional connectivity and influence behavior remain unexplored.

METHODS

We investigated the comprehensive environment-brain-behavior triple interactions through rigorous association, prediction, and mediation analyses, while adjusting for multiple confounders. Particularly, we examined the predictive power of brain functional network connectivity (FNC) and 41 environmental exposures for 23 behaviors related to cognitive ability and mental health in 7655 children selected from the Adolescent Brain Cognitive Development (ABCD) Study at both baseline and follow-up.

RESULTS

FNC demonstrated more predictability for cognitive abilities than for mental health, with cross-validation from the UK Biobank study (N = 20,852), highlighting the importance of thalamus and hippocampus in longitudinal prediction, while FNC+environment demonstrated more predictive power than FNC in both cross-sectional and longitudinal prediction of all behaviors, especially for mental health (r = 0.32-0.63). We found that family and neighborhood exposures were common critical environmental influencers on cognitive ability and mental health, which can be mediated by FNC significantly. Healthy perinatal development was a unique protective factor for higher cognitive ability, whereas sleep problems, family conflicts, and adverse school environments specifically increased risk of poor mental health.

CONCLUSIONS

This work revealed comprehensive environment-brain-behavior triple interactions based on the ABCD Study, identified cognitive control and default mode networks as the most predictive functional networks for a wide repertoire of behaviors, and underscored the long-lasting impact of critical environmental exposures on childhood development, in which sleep problems were the most prominent factors affecting mental health.

Searching Reproducible Brain Features using NeuroMark: Templates for Different Age Populations and Imaging Modalities

A primary challenge to the data-driven analysis is the balance between poor generalizability of population-based research and characterizing more subject-, study- and population-specific variability. We previously introduced a fully automated spatially constrained independent component analysis (ICA) framework called NeuroMark and its functional MRI (fMRI) template. NeuroMark has been successfully applied in numerous studies, identifying brain markers reproducible across datasets and disorders. The first NeuroMark template was constructed based on young adult cohorts. We recently expanded on this initiative by creating a standardized normative multi-spatial-scale functional template using over 100,000 subjects, aiming to improve generalizability and comparability across studies involving diverse cohorts. While a unified template across the lifespan is desirable, a comprehensive investigation of the similarities and differences between components from different age populations might help systematically transform our understanding of the human brain by revealing the most well-replicated and variable network features throughout the lifespan. In this work, we introduced two significant expansions of NeuroMark templates first by generating replicable fMRI templates for infants, adolescents, and aging cohorts, and second by incorporating structural MRI (sMRI) and diffusion MRI (dMRI) modalities. Specifically, we built spatiotemporal fMRI templates based on 6,000 resting-state scans from four datasets. This is the first attempt to create robust ICA templates covering dynamic brain development across the lifespan. For the sMRI and dMRI data, we used two large publicly available datasets including more than 30,000 scans to build reliable templates. We employed a spatial similarity analysis to identify replicable templates and investigate the degree to which unique and similar patterns are reflective in different age populations. Our results suggest remarkably high similarity of the resulting adapted components, even across extreme age differences. With the new templates, the NeuroMark framework allows us to perform age-specific adaptations and to capture features adaptable to each modality, therefore facilitating biomarker identification across brain disorders. In sum, the present work demonstrates the generalizability of NeuroMark templates and suggests the potential of new templates to boost accuracy in mental health research and advance our understanding of lifespan and cross-modal alterations.

Connectome-based prediction of decreased trust propensity in older adults with mild cognitive impairment: A resting-state functional magnetic resonance imaging study

Trust propensity (TP) relies more on social than economic rationality to transform the perceived probability of betrayal into positive reciprocity expectations in older adults with normal cognition. While deficits in social rationality have been observed in older adults with mild cognitive impairment (MCI), there is limited research on TP and its associated resting-state functional connectivity (RSFC) mechanisms in this population. To measure TP and related psychological functions (affect, motivation, executive cognition, and social cognition), MCI (n = 42) and normal healthy control (NHC, n = 115) groups completed a one-shot trust game and additional assessments of related psychological functions. RSFC associated with TP was analyzed using connectome-based predictive modeling (CPM) and lesion simulations. Our behavioral results showed that the MCI group trusted less (i.e., had lower TP) than the NHC group, with lower TP associated with higher sensitivity to the probability of betrayal in the MCI group. In the MCI group, only negative CPM models (RSFC negatively correlated with TP) significantly predicted TP, with a high salience network (SN) contribution. In contrast, in the NHC group, positive CPM models (RSFC positively correlated with TP) significantly predicted TP, with a high contribution from the default mode network (DMN). In addition, the total network strength of the NHC-specific positive network was lower in the MCI group than in the NHC group. Our findings demonstrated a decrease in TP in the MCI group compared to the NHC group, which is associated with deficits in social rationality (social cognition, associated with DMN) and increased sensitivity to betrayal (affect, associated with SN) in a trust dilemma. In conclusion, our study contributes to understanding MCI-related alterations in trust and their underlying neural mechanisms.

Predicting depression risk in early adolescence via multimodal brain imaging

Depression is an incapacitating psychiatric disorder with increased risk through adolescence. Among other factors, children with family history of depression have significantly higher risk of developing depression. Early identification of pre-adolescent children who are at risk of depression is crucial for early intervention and prevention. In this study, we used a large longitudinal sample from the Adolescent Brain Cognitive Development (ABCD) Study (2658 participants after imaging quality control, between 9-10 years at baseline), we applied advanced machine learning methods to predict depression risk at the two-year follow-up from the baseline assessment, using a set of comprehensive multimodal neuroimaging features derived from structural MRI, diffusion tensor imaging, and task and rest functional MRI. Prediction performance underwent a rigorous cross-validation method of leave-one-site-out. Our results demonstrate that all brain features had prediction scores significantly better than expected by chance, with brain features from rest-fMRI showing the best classification performance in the high-risk group of participants with parental history of depression (N = 625). Specifically, rest-fMRI features, which came from functional connectomes, showed significantly better classification performance than other brain features. This finding highlights the key role of the interacting elements of the connectome in capturing more individual variability in psychopathology compared to measures of single brain regions. Our study contributes to the effort of identifying biological risks of depression in early adolescence in population-based samples.

CI-GNN: A Granger causality-inspired graph neural network for interpretable brain network-based psychiatric diagnosis

There is a recent trend to leverage the power of graph neural networks (GNNs) for brain-network based psychiatric diagnosis, which, in turn, also motivates an urgent need for psychiatrists to fully understand the decision behavior of the used GNNs. However, most of the existing GNN explainers are either post-hoc in which another interpretive model needs to be created to explain a well-trained GNN, or do not consider the causal relationship between the extracted explanation and the decision, such that the explanation itself contains spurious correlations and suffers from weak faithfulness. In this work, we propose a granger causality-inspired graph neural network (CI-GNN), a built-in interpretable model that is able to identify the most influential subgraph (i.e., functional connectivity within brain regions) that is causally related to the decision (e.g., major depressive disorder patients or healthy controls), without the training of an auxillary interpretive network. CI-GNN learns disentangled subgraph-level representations α and β that encode, respectively, the causal and non-causal aspects of original graph under a graph variational autoencoder framework, regularized by a conditional mutual information (CMI) constraint. We theoretically justify the validity of the CMI regulation in capturing the causal relationship. We also empirically evaluate the performance of CI-GNN against three baseline GNNs and four state-of-the-art GNN explainers on synthetic data and three large-scale brain disease datasets. We observe that CI-GNN achieves the best performance in a wide range of metrics and provides more reliable and concise explanations which have clinical evidence. The source code and implementation details of CI-GNN are freely available at GitHub repository (https://github.com/ZKZ-Brain/CI-GNN/).

Maximum Classifier Discrepancy Generative Adversarial Network for Jointly Harmonizing Scanner Effects and Improving Reproducibility of Downstream Tasks

OBJECTIVE

Multi-site collaboration is essential for overcoming small-sample problems when exploring reproducible biomarkers in MRI studies. However, various scanner-specific factors dramatically reduce the cross-scanner replicability. Moreover, existing harmony methods mostly could not guarantee the improved performance of downstream tasks.

METHODS

we proposed a new multi-scanner harmony framework, called 'maximum classifier discrepancy generative adversarial network', or MCD-GAN, for removing scanner effects in the original feature space while preserving substantial biological information for downstream tasks. Specifically, the adversarial generative network was utilized for persisting the structural layout of each sample, and the maximum classifier discrepancy module was introduced for regulating GAN generators by incorporating the downstream tasks.

RESULTS

We compared the MCD-GAN with other state-of-the-art data harmony approaches (e.g., ComBat, CycleGAN) on simulated data and the Adolescent Brain Cognitive Development (ABCD) dataset. Results demonstrate that MCD-GAN outperformed other approaches in improving cross-scanner classification performance while preserving the anatomical layout of the original images.

SIGNIFICANCE

To the best of our knowledge, the proposed MCD-GAN is the first generative model which incorporates downstream tasks while harmonizing, and is a promising solution for facilitating cross-site reproducibility in various tasks such as classification and regression.

Co-representation of Functional Brain Networks Is Shaped by Cortical Myeloarchitecture and Reveals Individual Behavioral Ability

Large-scale functional networks are spatially distributed in the human brain. Despite recent progress in differentiating their functional roles, how the brain navigates the spatial coordination among them and the biological relevance of this coordination is still not fully understood. Capitalizing on canonical individualized networks derived from functional MRI data, we proposed a new concept, that is, co-representation of functional brain networks, to delineate the spatial coordination among them. To further quantify the co-representation pattern, we defined two indexes, that is, the co-representation specificity (CoRS) and intensity (CoRI), for separately measuring the extent of specific and average expression of functional networks at each brain location by using the data from both sexes. We found that the identified pattern of co-representation was anchored by cortical regions with three types of cytoarchitectural classes along a sensory-fugal axis, including, at the first end, primary (idiotypic) regions showing high CoRS, at the second end, heteromodal regions showing low CoRS and high CoRI, at the third end, paralimbic regions showing low CoRI. Importantly, we demonstrated the critical role of myeloarchitecture in sculpting the spatial distribution of co-representation by assessing the association with the myelin-related neuroanatomical and transcriptomic profiles. Furthermore, the significance of manifesting the co-representation was revealed in its prediction of individual behavioral ability. Our findings indicated that the spatial coordination among functional networks was built upon an anatomically configured blueprint to facilitate neural information processing, while advancing our understanding of the topographical organization of the brain by emphasizing the assembly of functional networks.

Cumulative Effects of Resting-State Connectivity Across All Brain Networks Significantly Correlate with Attention-Deficit Hyperactivity Disorder Symptoms

Identification of replicable neuroimaging correlates of attention-deficit hyperactivity disorder (ADHD) has been hindered by small sample sizes, small effects, and heterogeneity of methods. Given evidence that ADHD is associated with alterations in widely distributed brain networks and the small effects of individual brain features, a whole-brain perspective focusing on cumulative effects is warranted. The use of large, multisite samples is crucial for improving reproducibility and clinical utility of brain-wide MRI association studies. To address this, a polyneuro risk score (PNRS) representing cumulative, brain-wide, ADHD-associated resting-state functional connectivity was constructed and validated using data from the Adolescent Brain Cognitive Development (ABCD, N = 5,543, 51.5% female) study, and was further tested in the independent Oregon-ADHD-1000 case-control cohort (N = 553, 37.4% female). The ADHD PNRS was significantly associated with ADHD symptoms in both cohorts after accounting for relevant covariates (p < 0.001). The most predictive PNRS involved all brain networks, though the strongest effects were concentrated among the default mode and cingulo-opercular networks. In the longitudinal Oregon-ADHD-1000, non-ADHD youth had significantly lower PNRS (Cohen's d = -0.318, robust p = 5.5 × 10-4) than those with persistent ADHD (age 7-19). The PNRS, however, did not mediate polygenic risk for ADHD. Brain-wide connectivity was robustly associated with ADHD symptoms in two independent cohorts, providing further evidence of widespread dysconnectivity in ADHD. Evaluation in enriched samples demonstrates the promise of the PNRS approach for improving reproducibility in neuroimaging studies and unraveling the complex relationships between brain connectivity and behavioral disorders.

Structural brain feature is associated with changes of masticatory performance in healthy elderly people: Evidence from longitudinal neuroimaging research over two years

BACKGROUND

Human neuroimaging studies have revealed the association between brain structure and masticatory function. However, the majority of the studies adopted a cross-sectional design, which hardly reveals the change in masticatory function and brain structure between different timepoints, and the dynamical association between changes in masticatory function and changes in brain structure has not been elucidated.

OBJECTIVE

With a longitudinal design, we assessed the association between changes in masticatory performance (MP) and regional brain volume.

METHODS

Twenty-two elderly participants received assessments of the number of missing teeth and MP (via colour-changeable chewing gum) when they entered the study (i.e. the initial stage, T0 ), approximately 6 months later (T0.5 ), and approximately 1-2 years later (T1 ). Difficulty of food intake was assessed using a questionnaire. The participants received magnetic resonance imaging (MRI) at T0 and T1 . The brain volume of the motor-related area was estimated using FreeSurfer for MRI data. The associations between different stages were analysed using Spearman's rho correlation coefficients.

RESULTS

(1) Individually, a smaller volume of right primary motor cortex at T0 was associated with increased MP from T0 to T1 , suggesting the brain's role in changing oral functions; (2) higher MP at T0 was associated with an increased volume of the left superior frontal cortex from T0 to T1 p, suggesting a potential effect on brain plasticity, and (3) increased difficulty to eat was associated with decrease MP but not brain volume of motor-related area.

CONCLUSIONS

The preliminary findings revealed a complicated pattern of structural brain features and masticatory function in elderly people, and either the hypothesis that the brain predisposes masticatory function or the hypothesis that mastication reshapes the brain is oversimplified.

Functional network interactions in patients with schizophrenia with persistent auditory verbal hallucinations: A multimodal MRI fusion approach using three-way pICA

Over the last decade, there have been an increasing number of functional magnetic resonance imaging (fMRI) studies examining brain activity in schizophrenia (SZ) patients with persistent auditory verbal hallucinations (AVH) using either task-based or resting-state fMRI (rs-fMRI) paradigms. Such data have been conventionally collected and analyzed as distinct modalities, disregarding putative crossmodal interactions. Recently, it has become possible to incorporate two or more modalities in one comprehensive analysis to uncover hidden patterns of neural dysfunction not sufficiently captured by separate analysis. A novel multivariate fusion approach to multimodal data analysis, i.e., parallel independent component analysis (pICA), has been previously shown to be a powerful tool in this regard. We utilized three-way pICA to study covarying components among fractional amplitude of low-frequency fluctuations (fALFF) for rs-MRI and task-based activation computed from an alertness and a working memory (WM) paradigm of 15 SZ patients with AVH, 16 non-hallucinating SZ patients (nAVH), and 19 healthy controls (HC). The strongest connected triplet (false discovery rate (FDR)-corrected pairwise correlations) comprised a frontostriatal/temporal network (fALFF), a temporal/sensorimotor network (alertness task), and a frontoparietal network (WM task). Frontoparietal and frontostriatal/temporal network strength significantly differed between AVH patients and HC. Phenomenological features such as omnipotence and malevolence of AVH were associated with temporal/sensorimotor and frontoparietal network strength. The transmodal data confirm a complex interplay of neural systems subserving attentional processes and cognitive control interacting with speech and language processing networks. In addition, the data emphasize the importance of sensorimotor regions modulating specific symptom dimensions of AVH.

Data leakage inflates prediction performance in connectome-based machine learning models

Predictive modeling is a central technique in neuroimaging to identify brain-behavior relationships and test their generalizability to unseen data. However, data leakage undermines the validity of predictive models by breaching the separation between training and test data. Leakage is always an incorrect practice but still pervasive in machine learning. Understanding its effects on neuroimaging predictive models can inform how leakage affects existing literature. Here, we investigate the effects of five forms of leakage-involving feature selection, covariate correction, and dependence between subjects-on functional and structural connectome-based machine learning models across four datasets and three phenotypes. Leakage via feature selection and repeated subjects drastically inflates prediction performance, whereas other forms of leakage have minor effects. Furthermore, small datasets exacerbate the effects of leakage. Overall, our results illustrate the variable effects of leakage and underscore the importance of avoiding data leakage to improve the validity and reproducibility of predictive modeling.

How do tasks impact the reliability of fMRI functional connectivity?

While there is growing interest in the use of functional magnetic resonance imaging-functional connectivity (fMRI-FC) for biomarker research, low measurement reliability of conventional acquisitions may limit applications. Factors known to impact FC reliability include scan length, head motion, signal properties, such as temporal signal-to-noise ratio (tSNR), and the acquisition state or task. As tasks impact signal in a region-wise fashion, they likely impact FC reliability differently across the brain, making task an important decision in study design. Here, we use the densely sampled Midnight Scan Club (MSC) dataset, comprising 5 h of rest and 6 h of task fMRI data in 10 healthy adults, to investigate regional effects of tasks on FC reliability. We further considered how BOLD signal properties contributing to tSNR, that is, temporal mean signal (tMean) and temporal standard deviation (tSD), vary across the brain, associate with FC reliability, and are modulated by tasks. We found that, relative to rest, tasks enhanced FC reliability and increased tSD for specific task-engaged regions. However, FC signal variability and reliability is broadly dampened during tasks outside task-engaged regions. From our analyses, we observed signal variability was the strongest driver of FC reliability. Overall, our findings suggest that the choice of task can have an important impact on reliability and should be considered in relation to maximizing reliability in networks of interest as part of study design.

Functional magnetic resonance imaging in schizophrenia: current evidence, methodological advances, limitations and future directions

Functional neuroimaging emerged with great promise and has provided fundamental insights into the neurobiology of schizophrenia. However, it has faced challenges and criticisms, most notably a lack of clinical translation. This paper provides a comprehensive review and critical summary of the literature on functional neuroimaging, in particular functional magnetic resonance imaging (fMRI), in schizophrenia. We begin by reviewing research on fMRI biomarkers in schizophrenia and the clinical high risk phase through a historical lens, moving from case-control regional brain activation to global connectivity and advanced analytical approaches, and more recent machine learning algorithms to identify predictive neuroimaging features. Findings from fMRI studies of negative symptoms as well as of neurocognitive and social cognitive deficits are then reviewed. Functional neural markers of these symptoms and deficits may represent promising treatment targets in schizophrenia. Next, we summarize fMRI research related to antipsychotic medication, psychotherapy and psychosocial interventions, and neurostimulation, including treatment response and resistance, therapeutic mechanisms, and treatment targeting. We also review the utility of fMRI and data-driven approaches to dissect the heterogeneity of schizophrenia, moving beyond case-control comparisons, as well as methodological considerations and advances, including consortia and precision fMRI. Lastly, limitations and future directions of research in the field are discussed. Our comprehensive review suggests that, in order for fMRI to be clinically useful in the care of patients with schizophrenia, research should address potentially actionable clinical decisions that are routine in schizophrenia treatment, such as which antipsychotic should be prescribed or whether a given patient is likely to have persistent functional impairment. The potential clinical utility of fMRI is influenced by and must be weighed against cost and accessibility factors. Future evaluations of the utility of fMRI in prognostic and treatment response studies may consider including a health economics analysis.

Brain-phenotype predictions can survive across diverse real-world data

Recent work suggests that machine learning models predicting psychiatric treatment outcomes based on clinical data may fail when applied to unharmonized samples. Neuroimaging predictive models offer the opportunity to incorporate neurobiological information, which may be more robust to dataset shifts. Yet, among the minority of neuroimaging studies that undertake any form of external validation, there is a notable lack of attention to generalization across dataset-specific idiosyncrasies. Research settings, by design, remove the between-site variations that real-world and, eventually, clinical applications demand. Here, we rigorously test the ability of a range of predictive models to generalize across three diverse, unharmonized samples: the Philadelphia Neurodevelopmental Cohort (n=1291), the Healthy Brain Network (n=1110), and the Human Connectome Project in Development (n=428). These datasets have high inter-dataset heterogeneity, encompassing substantial variations in age distribution, sex, racial and ethnic minority representation, recruitment geography, clinical symptom burdens, fMRI tasks, sequences, and behavioral measures. We demonstrate that reproducible and generalizable brain-behavior associations can be realized across diverse dataset features with sample sizes in the hundreds. Results indicate the potential of functional connectivity-based predictive models to be robust despite substantial inter-dataset variability. Notably, for the HCPD and HBN datasets, the best predictions were not from training and testing in the same dataset (i.e., cross-validation) but across datasets. This result suggests that training on diverse data may improve prediction in specific cases. Overall, this work provides a critical foundation for future work evaluating the generalizability of neuroimaging predictive models in real-world scenarios and clinical settings.

Dysfunctions of multiscale dynamic brain functional networks in subjective cognitive decline

Subjective cognitive decline is potentially the earliest symptom of Alzheimer's disease, whose objective neurological basis remains elusive. To explore the potential biomarkers for subjective cognitive decline, we developed a novel deep learning method based on multiscale dynamical brain functional networks to identify subjective cognitive declines. We retrospectively constructed an internal data set (with 112 subjective cognitive decline and 64 healthy control subjects) to develop and internally validate the deep learning model. Conventional deep learning methods based on static and dynamic brain functional networks are compared. After the model is established, we prospectively collect an external data set (26 subjective cognitive decline and 12 healthy control subjects) for testing. Meanwhile, our method provides monitoring of the transitions between normal and abnormal (subjective cognitive decline-related) dynamical functional network states. The features of abnormal dynamical functional network states are quantified by network and variability metrics and associated with individual cognitions. Our method achieves an area under the receiver operating characteristic curve of 0.807 ± 0.046 in the internal validation data set and of 0.707 (P = 0.007) in the external testing data set, which shows improvements compared to conventional methods. The method further suggests that, at the local level, the abnormal dynamical functional network states are characterized by decreased connectivity strength and increased connectivity variability at different spatial scales. At the network level, the abnormal states are featured by scale-specifically altered modularity and all-scale decreased efficiency. Low tendencies to stay in abnormal states and high state transition variabilities are significantly associated with high general, language and executive functions. Overall, our work supports the deficits in multiscale brain dynamical functional networks detected by the deep learning method as reliable and meaningful neural alternation underpinning subjective cognitive decline.

Machine learning and artificial intelligence in neuroscience: A primer for researchers

Artificial intelligence (AI) is often used to describe the automation of complex tasks that we would attribute intelligence to. Machine learning (ML) is commonly understood as a set of methods used to develop an AI. Both have seen a recent boom in usage, both in scientific and commercial fields. For the scientific community, ML can solve bottle necks created by complex, multi-dimensional data generated, for example, by functional brain imaging or *omics approaches. ML can here identify patterns that could not have been found using traditional statistic approaches. However, ML comes with serious limitations that need to be kept in mind: their tendency to optimise solutions for the input data means it is of crucial importance to externally validate any findings before considering them more than a hypothesis. Their black-box nature implies that their decisions usually cannot be understood, which renders their use in medical decision making problematic and can lead to ethical issues. Here, we present an introduction for the curious to the field of ML/AI. We explain the principles as commonly used methods as well as recent methodological advancements before we discuss risks and what we see as future directions of the field. Finally, we show practical examples of neuroscience to illustrate the use and limitations of ML.

The effects of data leakage on connectome-based machine learning models

Predictive modeling has now become a central technique in neuroimaging to identify complex brain-behavior relationships and test their generalizability to unseen data. However, data leakage, which unintentionally breaches the separation between data used to train and test the model, undermines the validity of predictive models. Previous literature suggests that leakage is generally pervasive in machine learning, but few studies have empirically evaluated the effects of leakage in neuroimaging data. Although leakage is always an incorrect practice, understanding the effects of leakage on neuroimaging predictive models provides insight into the extent to which leakage may affect the literature. Here, we investigated the effects of leakage on machine learning models in two common neuroimaging modalities, functional and structural connectomes. Using over 400 different pipelines spanning four large datasets and three phenotypes, we evaluated five forms of leakage fitting into three broad categories: feature selection, covariate correction, and lack of independence between subjects. As expected, leakage via feature selection and repeated subjects drastically inflated prediction performance. Notably, other forms of leakage had only minor effects (e.g., leaky site correction) or even decreased prediction performance (e.g., leaky covariate regression). In some cases, leakage affected not only prediction performance, but also model coefficients, and thus neurobiological interpretations. Finally, we found that predictive models using small datasets were more sensitive to leakage. Overall, our results illustrate the variable effects of leakage on prediction pipelines and underscore the importance of avoiding data leakage to improve the validity and reproducibility of predictive modeling.

Sequence-Type Classification of Brain MRI for Acute Stroke Using a Self-Supervised Machine Learning Algorithm

We propose a self-supervised machine learning (ML) algorithm for sequence-type classification of brain MRI using a supervisory signal from DICOM metadata (i.e., a rule-based virtual label). A total of 1787 brain MRI datasets were constructed, including 1531 from hospitals and 256 from multi-center trial datasets. The ground truth (GT) was generated by two experienced image analysts and checked by a radiologist. An ML framework called ImageSort-net was developed using various features related to MRI acquisition parameters and used for training virtual labels and ML algorithms derived from rule-based labeling systems that act as labels for supervised learning. For the performance evaluation of ImageSort-net (MLvirtual), we compare and analyze the performances of models trained with human expert labels (MLhumans), using as a test set blank data that the rule-based labeling system failed to infer from each dataset. The performance of ImageSort-net (MLvirtual) was comparable to that of MLhuman (98.5% and 99%, respectively) in terms of overall accuracy when trained with hospital datasets. When trained with a relatively small multi-center trial dataset, the overall accuracy was relatively lower than that of MLhuman (95.6% and 99.4%, respectively). After integrating the two datasets and re-training them, MLvirtual showed higher accuracy than MLvirtual trained only on multi-center datasets (95.6% and 99.7%, respectively). Additionally, the multi-center dataset inference performances after the re-training of MLvirtual and MLhumans were identical (99.7%). Training of ML algorithms based on rule-based virtual labels achieved high accuracy for sequence-type classification of brain MRI and enabled us to build a sustainable self-learning system.

TSP-GNN: a novel neuropsychiatric disorder classification framework based on task-specific prior knowledge and graph neural network

Neuropsychiatric disorder (ND) is often accompanied by abnormal functional connectivity (FC) patterns in specific task contexts. The distinctive task-specific FC patterns can provide valuable features for ND classification models using deep learning. However, most previous studies rely solely on the whole-brain FC matrix without considering the prior knowledge of task-specific FC patterns. Insight by the decoding studies on brain-behavior relationship, we develop TSP-GNN, which extracts task-specific prior (TSP) connectome patterns and employs graph neural network (GNN) for disease classification. TSP-GNN was validated using publicly available datasets. Our results demonstrate that different ND types show distinct task-specific connectivity patterns. Compared with the whole-brain node characteristics, utilizing task-specific nodes enhances the accuracy of ND classification. TSP-GNN comprises the first attempt to incorporate prior task-specific connectome patterns and the power of deep learning. This study elucidates the association between brain dysfunction and specific cognitive processes, offering valuable insights into the cognitive mechanism of neuropsychiatric disease.

Oxytocin enhances the triangular association among behavior, resting-state, and task-state functional connectivity

Considerable advances in the role of oxytocin (OT) effect on behavior and the brain network have been made, but the effect of OT on the association between inter-individual differences in functional connectivity (FC) and behavior is elusive. Here, by using a face-perception task and multiple connectome-based predictive models, we aimed to (1) determine whether OT could enhance the association among behavioral performance, resting-state FC (rsFC), and task-state FC (tsFC) and (2) if so, explore the role of OT in enhancing this triangular association. We found that in the OT group, the prediction performance of using rsFC or tsFC to predict task behavior was higher than that of the PL group. Additionally, the correlation coefficient between rsFC and tsFC was substantially higher in the OT group than in the PL group. The strength of these associations could be partly explained by OT altering the brain's FCs related to social cognition and face perception in both the resting and task states, mainly in brain regions such as the limbic system, prefrontal cortex, temporal poles, and temporoparietal junction. Taken together, these results provide novel evidence and a corresponding mechanism for how neuropeptides cause increased associations among inter-individual differences across different levels (e.g., behavior and large-scale brain networks in both resting and task-state), and may inspire future research on the role of neuropeptides in the cross levels association of both clinical and nonclinical use.

Can we diagnose mental disorders in children? A large-scale assessment of machine learning on structural neuroimaging of 6916 children in the adolescent brain cognitive development study

Background

Prediction of mental disorders based on neuroimaging is an emerging area of research with promising first results in adults. However, research on the unique demographic of children is underrepresented and it is doubtful whether findings obtained on adults can be transferred to children.

Methods

Using data from 6916 children aged 9-10 in the multicenter Adolescent Brain Cognitive Development study, we extracted 136 regional volume and thickness measures from structural magnetic resonance images to rigorously evaluate the capabilities of machine learning to predict 10 different psychiatric disorders: major depressive disorder, bipolar disorder (BD), psychotic symptoms, attention deficit hyperactivity disorder (ADHD), oppositional defiant disorder, conduct disorder, post-traumatic stress disorder, obsessive-compulsive disorder, generalized anxiety disorder, and social anxiety disorder. For each disorder, we performed cross-validation and assessed whether models discovered a true pattern in the data via permutation testing.

Results

Two of 10 disorders can be detected with statistical significance when using advanced models that (i) allow for non-linear relationships between neuroanatomy and disorder, (ii) model interdependencies between disorders, and (iii) avoid confounding due to sociodemographic factors: ADHD (AUROC = 0.567, p = 0.002) and BD (AUROC = 0.551, p = 0.002). In contrast, traditional models perform consistently worse and predict only ADHD with statistical significance (AUROC = 0.529, p = 0.002).

Conclusion

While the modest absolute classification performance does not warrant application in the clinic, our results provide empirical evidence that embracing and explicitly accounting for the complexities of mental disorders via advanced machine learning models can discover patterns that would remain hidden with traditional models.

Chromatic fusion: Generative multimodal neuroimaging data fusion provides multi-informed insights into schizophrenia

This work proposes a novel generative multimodal approach to jointly analyze multimodal data while linking the multimodal information to colors. We apply our proposed framework, which disentangles multimodal data into private and shared sets of features from pairs of structural (sMRI), functional (sFNC and ICA), and diffusion MRI data (FA maps). With our approach, we find that heterogeneity in schizophrenia is potentially a function of modality pairs. Results show (1) schizophrenia is highly multimodal and includes changes in specific networks, (2) non-linear relationships with schizophrenia are observed when interpolating among shared latent dimensions, and (3) we observe a decrease in the modularity of functional connectivity and decreased visual-sensorimotor connectivity for schizophrenia patients for the FA-sFNC and sMRI-sFNC modality pairs, respectively. Additionally, our results generally indicate decreased fractional corpus callosum anisotropy, and decreased spatial ICA map and voxel-based morphometry strength in the superior frontal lobe as found in the FA-sFNC, sMRI-FA, and sMRI-ICA modality pair clusters. In sum, we introduce a powerful new multimodal neuroimaging framework designed to provide a rich and intuitive understanding of the data which we hope challenges the reader to think differently about how modalities interact.

The abilities of movie-watching functional connectivity in individual identifications and individualized predictions

Quite a few studies have been performed based on movie-watching functional connectivity (FC). As compared to its resting-state counterpart, however, there is still much to know about its abilities in individual identifications and individualized predictions. To pave the way for appropriate usage of movie-watching FC, we systemically evaluated the minimum number of time points, as well as the exact functional networks, supporting individual identifications and individualized predictions of apparent traits based on it. We performed the study based on the 7T movie-watching fMRI data included in the HCP S1200 Release, and took IQ as the test case for the prediction analyses. The results indicate that movie-watching FC based on only 15 time points can support successful individual identifications (99.47%), and the connectivity contributed more to identifications were much associated with higher-order cognitive processes (the secondary visual network, the frontoparietal network and the posterior multimodal network). For individualized predictions of IQ, it was found that successful predictions necessitated 60 time points (predicted vs. actual IQ correlation significant at P < 0.05, based on 5,000 permutations), and the prediction accuracy increased logarithmically with the number of time points used for connectivity calculation. Furthermore, the connectivity that contributed more to individual identifications exhibited the strongest prediction ability. Collectively, our findings demonstrate that movie-watching FC can capture rich information about human brain function, and its ability in individualized predictions depends heavily on the length of fMRI scans.

A neuromarker for deficit syndrome in schizophrenia from a combination of structural and functional magnetic resonance imaging

AIM

Deficit schizophrenia (DS), defined by primary and enduring negative symptoms, has been proposed as a promising homogeneous subtype of schizophrenia. It has been demonstrated that unimodal neuroimaging characteristics of DS were different from non-deficit schizophrenia (NDS), however, whether multimodal-based neuroimaging features could identify deficit syndrome remains to be determined.

METHODS

Functional and structural multimodal magnetic resonance imaging of DS, NDS and healthy controls were scanned. Voxel-based features of gray matter volume, fractional amplitude of low-frequency fluctuations, and regional homogeneity were extracted. The support vector machine classification models were constructed using these features separately and jointly. The most discriminative features were defined as the first 10% of features with the greatest weights. Moreover, relevance vector regression was applied to explore the predictive values of these top-weighted features in predicting negative symptoms.

RESULTS

The multimodal classifier achieved a higher accuracy (75.48%) compared with the single modal model in distinguishing DS from NDS. The most predictive brain regions were mainly located in the default mode and visual networks, exhibiting differences between functional and structural features. Further, the identified discriminative features significantly predicted scores of diminished expressivity factor in DS but not NDS.

CONCLUSIONS

The present study demonstrated that local properties of brain regions extracted from multimodal imaging data could distinguish DS from NDS with a machine learning-based approach and confirmed the relationship between distinctive features and the negative symptoms subdomain. These findings may improve the identification of potential neuroimaging signatures and improve the clinical assessment of the deficit syndrome.

Data-driven multimodal fusion: approaches and applications in psychiatric research

In the era of big data, where vast amounts of information are being generated and collected at an unprecedented rate, there is a pressing demand for innovative data-driven multi-modal fusion methods. These methods aim to integrate diverse neuroimaging perspectives to extract meaningful insights and attain a more comprehensive understanding of complex psychiatric disorders. However, analyzing each modality separately may only reveal partial insights or miss out on important correlations between different types of data. This is where data-driven multi-modal fusion techniques come into play. By combining information from multiple modalities in a synergistic manner, these methods enable us to uncover hidden patterns and relationships that would otherwise remain unnoticed. In this paper, we present an extensive overview of data-driven multimodal fusion approaches with or without prior information, with specific emphasis on canonical correlation analysis and independent component analysis. The applications of such fusion methods are wide-ranging and allow us to incorporate multiple factors such as genetics, environment, cognition, and treatment outcomes across various brain disorders. After summarizing the diverse neuropsychiatric magnetic resonance imaging fusion applications, we further discuss the emerging neuroimaging analyzing trends in big data, such as N-way multimodal fusion, deep learning approaches, and clinical translation. Overall, multimodal fusion emerges as an imperative approach providing valuable insights into the underlying neural basis of mental disorders, which can uncover subtle abnormalities or potential biomarkers that may benefit targeted treatments and personalized medical interventions.

Individual differences in the language task-evoked and resting-state functional networks

The resting state functional network is highly variable across individuals. However, inter-individual differences in functional networks evoked by language tasks and their comparison with resting state are still unclear. To address these two questions, we used T1 anatomical data and functional brain imaging data of resting state and a story comprehension task from the Human Connectome Project (HCP) to characterize functional network variability and investigate the uniqueness of the functional network in both task and resting states. We first demonstrated that intrinsic and task-induced functional networks exhibited remarkable differences across individuals, and language tasks can constrain inter-individual variability in the functional brain network. Furthermore, we found that the inter-individual variability of functional networks in two states was broadly consistent and spatially heterogeneous, with high-level association areas manifesting more significant variability than primary visual processing areas. Our results suggested that the functional network underlying language comprehension is unique at the individual level, and the inter-individual variability architecture of the functional network is broadly consistent in language task and resting state.

Pre-COVID brain network topology prospectively predicts social anxiety alterations during the COVID-19 pandemic

Background

Social anxiety (SA) is a negative emotional response that can lead to mental health issues, which some have experienced during the coronavirus disease 2019 (COVID-19) pandemic. Little attention has been given to the neurobiological mechanisms underlying inter-individual differences in SA alterations related to COVID-19. This study aims to identify neurofunctional markers of COVID-specific SA development.

Methods

110 healthy participants underwent resting-state magnetic resonance imaging and behavioral tests before the pandemic (T1, October 2019 to January 2020) and completed follow-up behavioral measurements during the pandemic (T2, February to May 2020). We constructed individual functional networks and used graph theoretical analysis to estimate their global and nodal topological properties, then used Pearson correlation and partial least squares correlations examine their associations with COVID-specific SA alterations.

Results

In terms of global network parameters, SA alterations (T2-T1) were negatively related to pre-pandemic brain small-worldness and normalized clustering coefficient. In terms of nodal network parameters, SA alterations were positively linked to a pronounced degree centrality pattern, encompassing both the high-level cognitive networks (dorsal attention network, cingulo-opercular task control network, default mode network, memory retrieval network, fronto-parietal task control network, and subcortical network) and low-level perceptual networks (sensory/somatomotor network, auditory network, and visual network). These findings were robust after controlling for pre-pandemic general anxiety, other stressful life events, and family socioeconomic status, as well as by treating SA alterations as categorical variables.

Conclusions

The individual functional network associated with SA alterations showed a disrupted topological organization with a more random state, which may shed light on the neurobiological basis of COVID-related SA changes at the network level.

Estimating dynamic individual coactivation patterns based on densely sampled resting-state fMRI data and utilizing it for better subject identification

As a complex dynamic system, the brain exhibits spatially organized recurring patterns of activity over time. Coactivation patterns (CAPs), which analyzes data from each single frame, have been utilized to detect transient brain activity states recently. However, previous CAP analyses have been conducted at the group level, which might neglect meaningful individual differences. Here, we estimated individual CAP states at both subject- and scan-level based on a densely sampled dataset: Midnight Scan Club. We used differential identifiability, which measures the gap between intra- and inter-subject similarity, to evaluate individual differences. We found individual CAPs at the subject-level achieved the best fingerprinting ability by maintaining high intra-subject similarity and enlarging inter-subject differences, and brain regions of association networks mainly contributed to the identifiability. On the other hand, scan-level CAP states were unstable across scans for the same participant. Expectedly, we found subject-specific CAPs became more reliable and discriminative with more data (i.e., longer duration). As the acquisition time of each participant is limited in practice, our results recommend a data collection strategy that collects more scans with appropriate duration (e.g., 12 ~ 15 min/scan) to obtain more reliable subject-specific CAPs, when total acquisition time is fixed (e.g., 150 min). In summary, this work has constructed reliable subject-specific CAP states with meaningful individual differences, and recommended an appropriate data collection strategy, which can guide subsequent investigations into individualized brain dynamics.

Cross Atlas Remapping via Optimal Transport (CAROT): Creating connectomes for different atlases when raw data is not available

Open-source, publicly available neuroimaging datasets - whether from large-scale data collection efforts or pooled from multiple smaller studies - offer unprecedented sample sizes and promote generalization efforts. Releasing data can democratize science, increase the replicability of findings, and lead to discoveries. Partly due to patient privacy, computational, and data storage concerns, researchers typically release preprocessed data with the voxelwise time series parcellated into a map of predefined regions, known as an atlas. However, releasing preprocessed data also limits the choices available to the end-user. This is especially true for connectomics, as connectomes created from different atlases are not directly comparable. Since there exist several atlases with no gold standards, it is unrealistic to have processed, open-source data available from all atlases. Together, these limitations directly inhibit the potential benefits of open-source neuroimaging data. To address these limitations, we introduce Cross Atlas Remapping via Optimal Transport (CAROT) to find a mapping between two atlases. This approach allows data processed from one atlas to be directly transformed into a connectome based on another atlas without the need for raw data access. To validate CAROT, we compare reconstructed connectomes against their original counterparts (i.e., connectomes generated directly from an atlas), demonstrate the utility of transformed connectomes in downstream analyses, and show how a connectome-based predictive model can generalize to publicly available data that was processed with different atlases. Overall, CAROT can reconstruct connectomes from an extensive set of atlases - without needing the raw data - allowing already processed connectomes to be easily reused in a wide range of analyses while eliminating redundant processing efforts. We share this tool as both source code and as a stand-alone web application (http://carotproject.com/).

Accurate Prediction of Momentary Cognition From Intensive Longitudinal Data

BACKGROUND

Deficits in cognitive performance are implicated in the development and maintenance of psychopathology. Emerging evidence further suggests that within-person fluctuations in cognitive performance may represent sensitive early markers of neuropsychiatric decline. Incorporating routine cognitive assessments into standard clinical care-to identify between-person differences and monitor within-person fluctuations-has the potential to improve diagnostic screening and treatment planning. In support of these goals, it is critical to understand to what extent cognitive performance varies under routine, remote assessment conditions (i.e., momentary cognition) in relation to a wide range of possible predictors.

METHODS

Using data-driven, high-dimensional methods, we ranked strong predictors of momentary cognition and evaluated out-of-sample predictive accuracy. Our approach leveraged innovations in digital technology, including ambulatory assessment of cognition and behavior 1) at scale (n = 122 participants, n = 94 females), 2) in naturalistic environments, and 3) within an intensive longitudinal study design (mean = 25.5 assessments/participant).

RESULTS

Reaction time (R2 > 0.70) and accuracy (0.56 >R2 > 0.35) were strongly predicted by age, between-person differences in mean performance, and time of day. Effects of self-reported, intraindividual fluctuations in environmental (e.g., noise) and internal (e.g., stress) states were also observed.

CONCLUSIONS

Our results provide robust estimates of effect size to characterize sources of cognitive variability, to support the identification of optimal windows for psychosocial interventions, and to possibly inform clinical evaluation under remote neuropsychological assessment conditions.

A meta-analysis and systematic review of single vs. multimodal neuroimaging techniques in the classification of psychosis

BACKGROUND

Psychotic disorders are characterized by structural and functional abnormalities in brain networks. Neuroimaging techniques map and characterize such abnormalities using unique features (e.g., structural integrity, coactivation). However, it is unclear if a specific method, or a combination of modalities, is particularly effective in identifying differences in brain networks of someone with a psychotic disorder.

METHODS

A systematic meta-analysis evaluated machine learning classification of schizophrenia spectrum disorders in comparison to healthy control participants using various neuroimaging modalities (i.e., T1-weighted imaging (T1), diffusion tensor imaging (DTI), resting state functional connectivity (rs-FC), or some combination (multimodal)). Criteria for manuscript inclusion included whole-brain analyses and cross-validation to provide a complete picture regarding the predictive ability of large-scale brain systems in psychosis. For this meta-analysis, we searched Ovid MEDLINE, PubMed, PsychInfo, Google Scholar, and Web of Science published between inception and March 13th 2023. Prediction results were averaged for studies using the same dataset, but parallel analyses were run that included studies with pooled sample across many datasets. We assessed bias through funnel plot asymmetry. A bivariate regression model determined whether differences in imaging modality, demographics, and preprocessing methods moderated classification. Separate models were run for studies with internal prediction (via cross-validation) and external prediction.

RESULTS

93 studies were identified for quantitative review (30 T1, 9 DTI, 40 rs-FC, and 14 multimodal). As a whole, all modalities reliably differentiated those with schizophrenia spectrum disorders from controls (OR = 2.64 (95%CI = 2.33 to 2.95)). However, classification was relatively similar across modalities: no differences were seen across modalities in the classification of independent internal data, and a small advantage was seen for rs-FC studies relative to T1 studies in classification in external datasets. We found large amounts of heterogeneity across results resulting in significant signs of bias in funnel plots and Egger's tests. Results remained similar, however, when studies were restricted to those with less heterogeneity, with continued small advantages for rs-FC relative to structural measures. Notably, in all cases, no significant differences were seen between multimodal and unimodal approaches, with rs-FC and unimodal studies reporting largely overlapping classification performance. Differences in demographics and analysis or denoising were not associated with changes in classification scores.

CONCLUSIONS

The results of this study suggest that neuroimaging approaches have promise in the classification of psychosis. Interestingly, at present most modalities perform similarly in the classification of psychosis, with slight advantages for rs-FC relative to structural modalities in some specific cases. Notably, results differed substantially across studies, with suggestions of biased effect sizes, particularly highlighting the need for more studies using external prediction and large sample sizes. Adopting more rigorous and systematized standards will add significant value toward understanding and treating this critical population.

The challenges and prospects of brain-based prediction of behaviour

Relating individual brain patterns to behaviour is fundamental in system neuroscience. Recently, the predictive modelling approach has become increasingly popular, largely due to the recent availability of large open datasets and access to computational resources. This means that we can use machine learning models and interindividual differences at the brain level represented by neuroimaging features to predict interindividual differences in behavioural measures. By doing so, we could identify biomarkers and neural correlates in a data-driven fashion. Nevertheless, this budding field of neuroimaging-based predictive modelling is facing issues that may limit its potential applications. Here we review these existing challenges, as well as those that we anticipate as the field develops. We focus on the impacts of these challenges on brain-based predictions. We suggest potential solutions to address the resolvable challenges, while keeping in mind that some general and conceptual limitations may also underlie the predictive modelling approach.

Multi-modal MRI measures reveal sensory abnormalities in major depressive disorder patients: A surface-based study

BACKGROUND

Multi-modal magnetic resonance imaging (MRI) measures are supposed to be able to capture different brain neurobiological aspects of major depressive disorder (MDD). A fusion analysis of structural and functional modalities may better reveal the disease biomarker specific to the MDD disease.

METHODS

We recruited 30 MDD patients and 30 matched healthy controls (HC). For each subject, we acquired high-resolution brain structural images and resting-state fMRI (rs-fMRI) data using a 3 T MRI scanner. We first extracted the brain morphometric measures, including the cortical volume (CV), cortical thickness (CT), and surface area (SA), for each subject from the structural images, and then detected the structural clusters showing significant between-group differences in each measure using the surface-based morphology (SBM) analysis. By taking the identified structural clusters as seeds, we performed seed-based functional connectivity (FC) analyses to determine the regions with abnormal FC in the patients. Based on a logistic regression model, we performed a classification analysis by selecting these structural and functional cluster-wise measures as features to distinguish the MDD patients from the HC.

RESULTS

The MDD patients showed significantly lower CV in a cluster involving the right superior temporal gyrus (STG) and middle temporal gyrus (MTG), and lower SA in three clusters involving the bilateral STG, temporal pole gyrus, and entorhinal cortex, and the left inferior temporal gyrus, and fusiform gyrus, than the controls. No significant difference in CT was detected between the two groups. By taking the above-detected clusters as seeds to perform the seed-based FC analysis, we found that the MDD patients showed significantly lower FC between STG/MTG (CV's cluster) and two clusters located in the bilateral visual cortices than the controls. The logistic regression model based on the structural and functional features reached a classification accuracy of 86.7% (p < 0.001) between MDD and controls.

CONCLUSION

The present study showed sensory abnormalities in MDD patients using the multi-modal MRI analysis. This finding may act as a disease biomarker distinguishing MDD patients from healthy individuals.

Prediction of cognitive scores by joint use of movie-watching fMRI connectivity and eye tracking via Attention-CensNet

Background

Brain functional connectivity under the naturalistic paradigm has been shown to be better at predicting individual behaviors than other brain states, such as rest and doing tasks. Nevertheless, the state-of-the-art methods have found it difficult to achieve desirable results from movie-watching paradigm functional magnetic resonance imaging (mfMRI) -induced brain functional connectivity, especially when there are fewer datasets. Incorporating other physical measurements into the prediction method may enhance accuracy. Eye tracking, becoming popular due to its portability and lower expense, can provide abundant behavioral features related to the output of human's cognition, and thus might supplement the mfMRI in observing participants' subconscious behaviors. However, there are very few studies on how to effectively integrate the multimodal information to strengthen the performance by a unified framework.

Objective

A fusion approach with mfMRI and eye tracking, based on convolution with edge-node switching in graph neural networks (CensNet), is proposed in this article.

Methods

In this graph model, participants are designated as nodes, mfMRI derived functional connectivity as node features, and different eye-tracking features are used to compute similarity between participants to construct heterogeneous graph edges. By taking multiple graphs as different channels, we introduce squeeze-and-excitation attention module to CensNet (A-CensNet) to integrate graph embeddings from multiple channels into one.

Results

The proposed model outperforms those using a single modality and single channel, and state-of-the-art methods.

Conclusions

The results indicate that brain functional activities and eye behaviors might complement each other in interpreting trait-like phenotypes.

Sampling inequalities affect generalization of neuroimaging-based diagnostic classifiers in psychiatry

BACKGROUND

The development of machine learning models for aiding in the diagnosis of mental disorder is recognized as a significant breakthrough in the field of psychiatry. However, clinical practice of such models remains a challenge, with poor generalizability being a major limitation.

METHODS

Here, we conducted a pre-registered meta-research assessment on neuroimaging-based models in the psychiatric literature, quantitatively examining global and regional sampling issues over recent decades, from a view that has been relatively underexplored. A total of 476 studies (n = 118,137) were included in the current assessment. Based on these findings, we built a comprehensive 5-star rating system to quantitatively evaluate the quality of existing machine learning models for psychiatric diagnoses.

RESULTS

A global sampling inequality in these models was revealed quantitatively (sampling Gini coefficient (G) = 0.81, p < .01), varying across different countries (regions) (e.g., China, G = 0.47; the USA, G = 0.58; Germany, G = 0.78; the UK, G = 0.87). Furthermore, the severity of this sampling inequality was significantly predicted by national economic levels (β =  - 2.75, p < .001, R²_adj = 0.40; r =  - .84, 95% CI: - .41 to - .97), and was plausibly predictable for model performance, with higher sampling inequality for reporting higher classification accuracy. Further analyses showed that lack of independent testing (84.24% of models, 95% CI: 81.0-87.5%), improper cross-validation (51.68% of models, 95% CI: 47.2-56.2%), and poor technical transparency (87.8% of models, 95% CI: 84.9-90.8%)/availability (80.88% of models, 95% CI: 77.3-84.4%) are prevailing in current diagnostic classifiers despite improvements over time. Relating to these observations, model performances were found decreased in studies with independent cross-country sampling validations (all p < .001, BF₁₀ > 15). In light of this, we proposed a purpose-built quantitative assessment checklist, which demonstrated that the overall ratings of these models increased by publication year but were negatively associated with model performance.

CONCLUSIONS

Together, improving sampling economic equality and hence the quality of machine learning models may be a crucial facet to plausibly translating neuroimaging-based diagnostic classifiers into clinical practice.