Predicting Individual Task Contrasts From Resting-state Functional Connectivity using a Surface-based Convolutional Network

Task-Rest Reconfiguration Efficiency of the Reward Network Across Adolescence and Its Association With Early Life Stress and Depressive Symptoms

OBJECTIVE

Adolescents face significant changes in many domains of their daily lives that require them to flexibly adapt to changing environmental demands. To shift efficiently among various goals, adolescents must reconfigure their brains, disengaging from previous tasks and engaging in new activities.

METHOD

To examine this reconfiguration, we obtained resting-state and task-based functional magnetic resonance imaging (fMRI) scans in a community sample of 164 youths. We assessed the similarity of functional connectivity (FC) of the reward network between resting state and a reward-processing state, indexing the degree of reward network reconfiguration required to meet task demands. Given research documenting relations among reward network function, early life stress (ELS), and adolescent depression, we examined the association of reconfiguration efficiency with age across adolescence, the moderating effect of ELS on this association, and the relation between reconfiguration efficiency and depressive symptoms.

RESULTS

We found that older adolescents showed greater reconfiguration efficiency than younger adolescents and, furthermore, that this age-related association was moderated by the experience of ELS.

CONCLUSION

These findings suggest that reconfiguration efficiency of the reward network increases over adolescence, a developmental pattern that is attenuated in adolescents exposed to severe ELS. In addition, even after controlling for the effects of age and exposure to ELS, adolescents with higher levels of depressive symptoms exhibited greater reconfiguration efficiency, suggesting that they have brain states at rest that are more strongly optimized for reward processing than do asymptomatic youth.

PLAIN LANGUAGE SUMMARY

Adolescents face significant changes in many domains of their lives which requires them to flexibly adapt and reconfigure their brains to disengage from previous tasks and engage in new activities. In this study of a sample of 164 youth aged 9 to 20, the authors found an age-related increase in the reconfiguration efficiency of the reward network, which was pronounced in older adolescents exposed to severe early life stress. In addition, the study findings indicate that adolescents with higher levels of depressive symptoms showed greater reconfiguration efficiency, suggesting that their brains may be more optimized for processing rewards even at rest compared to their peers without any symptoms.

DIVERSITY & INCLUSION STATEMENT

We worked to ensure race, ethnic, and/or other types of diversity in the recruitment of human participants. We worked to ensure sex and gender balance in the recruitment of human participants. We worked to ensure that the study questionnaires were prepared in an inclusive way. One or more of the authors of this paper self-identifies as a member of one or more historically underrepresented racial and/or ethnic groups in science.

Temporal dysregulation of the somatomotor network in agitated depression

Agitated depression (A-MDD) is a severe subtype of major depressive disorder, with an increased risk of suicidality and the potential to evolve into bipolar disorder. Despite its clinical significance, the neural basis remains unclear. We hypothesize that psychomotor agitation, marked by pressured speech and racing thoughts, is linked to disruptions in brain dynamics. To test this hypothesis, we examined brain dynamics using time delay estimation and edge-centre time series, as well as dynamic connections between the somatomotor network (SMN) and the default mode network in 44 patients with A-MDD, 75 with non-agitated MDD (NA-MDD), and 94 healthy controls. Our results revealed that the neural co-activity duration was shorter in the A-MDD group compared with both the NA-MDD and controls (A-MDD versus NA-MDD: t = 2.295; A-MDD versus controls: t = 2.192, all P < 0.05). In addition, the dynamic of neural fluctuation in SMN altered in the A-MDD group than in the NA-MDD group (t = -2.616, P = 0.011) and was correlated with agitation severity (β = -0.228, P = 0.011). The inter-network connection was reduced in the A-MDD group compared with the control group (t = 2.102, P = 0.037), especially at low-amplitude time points (t = 2.139, P = 0.034). These findings indicate rapid neural fluctuations and disrupted dynamic coupling between the SMN and default mode network in A-MDD, potentially underlying the psychomotor agitation characteristic of this subtype. These insights contribute to a more nuanced understanding of the heterogeneity of depression and have implications for differential diagnosis and treatment strategies.

An Adaptively Weighted Averaging Method for Regional Time Series Extraction of fMRI-Based Brain Decoding

Brain decoding that classifies cognitive states using the functional fluctuations of the brain can provide insightful information for understanding the brain mechanisms of cognitive functions. Among the common procedures of decoding the brain cognitive states with functional magnetic resonance imaging (fMRI), extracting the time series of each brain region after brain parcellation traditionally averages across the voxels within a brain region. This neglects the spatial information among the voxels and the requirement of extracting information for the downstream tasks. In this study, we propose to use a fully connected neural network that is jointly trained with the brain decoder to perform an adaptively weighted average across the voxels within each brain region. We perform extensive evaluations by cognitive state decoding, manifold learning, and interpretability analysis on the Human Connectome Project (HCP) dataset. The performance comparison of the cognitive state decoding presents an accuracy increase of up to 5% and stable accuracy improvement under different time window sizes, resampling sizes, and training data sizes. The results of manifold learning show that our method presents a considerable separability among cognitive states and basically excludes subject-specific information. The interpretability analysis shows that our method can identify reasonable brain regions corresponding to each cognitive state. Our study would aid the improvement of the basic pipeline of fMRI processing.

TS-AI: A deep learning pipeline for multimodal subject-specific parcellation with task contrasts synthesis

Accurate mapping of brain functional subregions at an individual level is crucial. Task-based functional MRI (tfMRI) captures subject-specific activation patterns during various functions and behaviors, facilitating the individual localization of functionally distinct subregions. However, acquiring high-quality tfMRI is time-consuming and resource-intensive in both scientific and clinical settings. The present study proposes a two-stage network model, TS-AI, to individualize an atlas on cortical surfaces through the prediction of tfMRI data. TS-AI first synthesizes a battery of task contrast maps for each individual by leveraging tract-wise anatomical connectivity and resting-state networks. These synthesized maps, along with feature maps of tract-wise anatomical connectivity and resting-state networks, are then fed into an end-to-end deep neural network to individualize an atlas. TS-AI enables the synthesized task contrast maps to be used in individual parcellation without the acquisition of actual task fMRI scans. In addition, a novel feature consistency loss is designed to assign vertices with similar features to the same parcel, which increases individual specificity and mitigates overfitting risks caused by the absence of individual parcellation ground truth. The individualized parcellations were validated by assessing test-retest reliability, homogeneity, and cognitive behavior prediction using diverse reference atlases and datasets, demonstrating the superior performance and generalizability of TS-AI. Sensitivity analysis yielded insights into region-specific features influencing individual variation in functional regionalization. Additionally, TS-AI identified accelerated shrinkage in the medial temporal and cingulate parcels during the progression of Alzheimer's disease, suggesting its potential in clinical research and applications.

Predicting Task Activation Maps from Resting-State Functional Connectivity using Deep Learning

Recent work has shown that deep learning is a powerful tool for predicting brain activation patterns evoked through various tasks using resting state features. We replicate and improve upon this recent work to introduce two models, BrainSERF and BrainSurfGCN, that perform at least as well as the state-of-the-art while greatly reducing memory and computational footprints. Our performance analysis observed that low predictability was associated with a possible lack of task engagement derived from behavioral performance. Furthermore, a deficiency in model performance was also observed for closely matched task contrasts, likely due to high individual variability confirmed by low test-retest reliability. Overall, we successfully replicate recently developed deep learning architecture and provide scalable models for further research.

The Prediction of Brain Activity from Connectivity: Advances and Applications

The human brain is composed of multiple, discrete, functionally specialized regions that are interconnected to form large-scale distributed networks. Using advanced brain-imaging methods and machine-learning analytical approaches, recent studies have demonstrated that regional brain activity during the performance of various cognitive tasks can be accurately predicted from patterns of task-independent brain connectivity. In this review article, we first present evidence for the predictability of brain activity from structural connectivity (i.e., white matter connections) and functional connectivity (i.e., temporally synchronized task-free activations). We then discuss the implications of such predictions to clinical populations, such as patients diagnosed with psychiatric disorders or neurologic diseases, and to the study of brain-behavior associations. We conclude that connectivity may serve as an infrastructure that dictates brain activity, and we pinpoint several open questions and directions for future research.

Predicting high-level visual areas in the absence of task fMRI

The ventral visual stream is organized into units, or functional regions of interest (fROIs), specialized for processing high-level visual categories. Task-based fMRI scans ("localizers") are typically used to identify each individual's nuanced set of fROIs. The unique landscape of an individual's functional activation may rely in large part on their specialized connectivity patterns; recent studies corroborate this by showing that connectivity can predict individual differences in neural responses. We focus on the ventral visual stream and ask: how well can an individual's resting state functional connectivity localize their fROIs for face, body, scene, and object perception? And are the neural processors for any particular visual category better predicted by connectivity than others, suggesting a tighter mechanistic relationship between connectivity and function? We found, among 18 fROIs predicted from connectivity for each subject, all but one were selective for their preferred visual category. Defining an individual's fROIs based on their connectivity patterns yielded regions that were more selective than regions identified from previous studies or atlases in nearly all cases. Overall, we found that in the absence of a domain-specific localizer task, a 10-min resting state scan can be reliably used for defining these fROIs.

Elevated VCAM-1 levels in peripheral blood are associated with brain structural and functional alterations in major depressive disorder

BACKGROUND

Vascular cell adhesion molecule-1 (VCAM-1) is a well-known biomarker of endothelial activation. This study aimed to determine whether changes in peripheral VCAM-1 levels occurred in major depressive disorder (MDD) patients and explored immune-brain interactions based on neuroimaging.

METHODS

This study included 165 subjects (80 healthy controls [HCs] and 85 MDD patients). Of them, 133 underwent magnetic resonance imaging. VCAM-1 was measured using a commercially available Enzyme-Linked Immunosorbent Assay kit following the manufacturer's instructions. The gray matter volume (GMV) and surface-based functional connectivity (FC) were calculated based on Schaefer parcellation 400 parcels.

RESULTS

Compared with the HCs, MDD patients exhibited significantly higher level of VCAM-1. The correlation analysis showed that VCAM-1 had a significant negative correlation with GMV of the right medial frontal cortex (MFC) and postcentral (PostCG). The mediation analyses showed that VCAM-1 mediated the association between group and GMV of PostCG and the FC of left ventral prefrontal cortex (vPFC) with right inferior parietal lobe (IPL).

CONCLUSIONS

This study showed that a high level of VCAM-1 was associated to the decreased GMV in the right MFC and PostCG, and mediated the FC of the left vPFC with right IPL. These findings suggested that VCAM-1 might contribute to the etiology of MDD by influencing brain structure and function.

LIMITATIONS

The cross-sectional design makes it difficult to determine the causal relationship and dynamic effect among VCAM-1, brain structure/function features, and depressive symptoms.

Connectome-based schizophrenia prediction using structural connectivity - Deep Graph Neural Network(sc-DGNN)

BACKGROUND

Connectome is understanding the complex organization of the human brain's structural and functional connectivity is essential for gaining insights into cognitive processes and disorders.

OBJECTIVE

To improve the prediction accuracy of brain disorder issues, the current study investigates dysconnected subnetworks and graph structures associated with schizophrenia.

METHOD

By using the proposed structural connectivity-deep graph neural network (sc-DGNN) model and compared with machine learning (ML) and deep learning (DL) models.This work attempts to focus on eighty-eight subjects of diffusion magnetic resonance imaging (dMRI), three classical ML, and five DL models.

RESULT

The structural connectivity-deep graph neural network (sc-DGNN) model is proposed to effectively predict dysconnectedness associated with schizophrenia and exhibits superior performance compared to traditional ML and DL (GNNs) methods in terms of accuracy, sensitivity, specificity, precision, F1-score, and Area under receiver operating characteristic (AUC).

CONCLUSION

The classification task on schizophrenia using structural connectivity matrices and experimental results showed that linear discriminant analysis (LDA) performed 72% accuracy rate in ML models and sc-DGNN performed at a 93% accuracy rate in DL models to distinguish between schizophrenia and healthy patients.

Causally informed activity flow models provide mechanistic insight into network-generated cognitive activations

Brain activity flow models estimate the movement of task-evoked activity over brain connections to help explain network-generated task functionality. Activity flow models have been shown to accurately generate task-evoked brain activations across a wide variety of brain regions and task conditions. However, these models have had limited explanatory power, given known issues with causal interpretations of the standard functional connectivity measures used to parameterize activity flow models. We show here that functional/effective connectivity (FC) measures grounded in causal principles facilitate mechanistic interpretation of activity flow models. We progress from simple to complex FC measures, with each adding algorithmic details reflecting causal principles. This reflects many neuroscientists' preference for reduced FC measure complexity (to minimize assumptions, minimize compute time, and fully comprehend and easily communicate methodological details), which potentially trades off with causal validity. We start with Pearson correlation (the current field standard) to remain maximally relevant to the field, estimating causal validity across a range of FC measures using simulations and empirical fMRI data. Finally, we apply causal-FC-based activity flow modeling to a dorsolateral prefrontal cortex region (DLPFC), demonstrating distributed causal network mechanisms contributing to its strong activation during a working memory task. Notably, this fully distributed model is able to account for DLPFC working memory effects traditionally thought to rely primarily on within-region (i.e., not distributed) recurrent processes. Together, these results reveal the promise of parameterizing activity flow models using causal FC methods to identify network mechanisms underlying cognitive computations in the human brain.

Generalizing prediction of task-evoked brain activity across datasets and populations

Predictions of task-based functional magnetic resonance imaging (fMRI) from task-free resting-state (rs) fMRI have gained popularity over the past decade. This method holds a great promise for studying individual variability in brain function without the need to perform highly demanding tasks. However, in order to be broadly used, prediction models must prove to generalize beyond the dataset they were trained on. In this work, we test the generalizability of prediction of task-fMRI from rs-fMRI across sites, MRI vendors and age-groups. Moreover, we investigate the data requirements for successful prediction. We use the Human Connectome Project (HCP) dataset to explore how different combinations of training sample sizes and number of fMRI datapoints affect prediction success in various cognitive tasks. We then apply models trained on HCP data to predict brain activations in data from a different site, a different MRI vendor (Phillips vs. Siemens scanners) and a different age group (children from the HCP-development project). We demonstrate that, depending on the task, a training set of approximately 20 participants with 100 fMRI timepoints each yields the largest gain in model performance. Nevertheless, further increasing sample size and number of timepoints results in significantly improved predictions, until reaching approximately 450-600 training participants and 800-1000 timepoints. Overall, the number of fMRI timepoints influences prediction success more than the sample size. We further show that models trained on adequate amounts of data successfully generalize across sites, vendors and age groups and provide predictions that are both accurate and individual-specific. These findings suggest that large-scale publicly available datasets may be utilized to study brain function in smaller, unique samples.

Localized activity alternations in periventricular nodular heterotopia-related epilepsy

OBJECTIVE

Periventricular nodular heterotopia (PNH) is a common type of heterotopia usually characterized by epilepsy. Previous studies have identified alterations in structural and functional connectivity related to this disorder, but its local functional neural basis has received less attention. The purpose of this study was to combine univariate analysis and a Gaussian process classifier (GPC) to assess local activity and further explore neuropathological mechanisms in PNH-related epilepsy.

METHODS

We used a 3.0-T scanner to acquire resting-state data and measure local regional homogeneity (ReHo) alterations in 38 patients with PNH-related epilepsy and 38 healthy controls (HCs). We first assessed ReHo alterations by comparing the PNH group to the HC group using traditional univariate analysis. Next, we applied a GPC to explore whether ReHo could be used to differentiate PNH patients from healthy patients at an individual level.

RESULTS

Compared to HCs, PNH-related epilepsy patients exhibited lower ReHo in the left insula extending to the putamen as well as in the subgenual anterior cingulate cortex (sgACC) extending to the orbitofrontal cortex (OFC) [p < 0.05, family-wise error corrected]. Both of these regions were also correlated with epilepsy duration. Furthermore, the ReHo GPC classification yielded a 76.32% accuracy (sensitivity = 71.05% and specificity = 81.58%) with p < 0.001 after permutation testing.

INTERPRETATION

Using the resting-state approach, we identified localized activity alterations in the left insula extending to the putamen and the sgACC extending to the OFC, providing pathophysiological evidence of PNH. These local connectivity patterns may provide a means to differentiate PNH patients from HCs.

From sMRI to task-fMRI: A unified geometric deep learning framework for cross-modal brain anatomo-functional mapping

Achieving predictions of brain functional activation patterns/task-fMRI maps from its underlying anatomy is an important yet challenging problem. Once successful, it will not only open up new ways to understand how brain anatomy influences functional organization of the brain, but also provide new technical support for the clinical use of anatomical information to guide the localization of cortical functional areas. However, due to the non-Euclidean complex architecture of brain anatomy and the inherent low signal-to-noise ratio (SNR) properties of fMRI signals, the key challenge in building such a cross-modal brain anatomo-functional mapping is how to effectively learn the context-aware information of brain anatomy and overcome the interference of noise-containing task-fMRI labels on the learning process. In this work, we propose a Unified Geometric Deep Learning framework (BrainUGDL) to perform the cross-modal brain anatomo-functional mapping task. Considering that both global and local structures of brain anatomy have an impact on brain functions from their respective perspectives, we innovatively propose the novel Global Graph Encoding (GGE) unit and Local Graph Attention (LGA) unit embedded into two parallel branches, focusing on learning the high-level global and local context information, respectively. Specifically, GGE learns the global context information of each mesh vertex by building and encoding global interactions, and LGA learns the local context information of each mesh vertex by selectively aggregating patch structure enhanced features from its spatial neighbors. The information learnt from the two branches is then fused to form a comprehensive representation of brain anatomical features for final brain function predictions. To address the inevitable measurement noise in task-fMRI labels, we further elaborate a novel uncertainty-filtered learning mechanism, which enables BrainUGDL to realize revised learning from the noise-containing labels through the estimated uncertainty. Experiments across seven open task-fMRI datasets from human connectome project (HCP) demonstrate the superiority of BrainUGDL. To our best knowledge, our proposed BrainUGDL is the first to achieve the prediction of individual task-fMRI maps solely based on brain sMRI data.

Deep learning in neuroimaging data analysis: Applications, challenges, and solutions

Methods for the analysis of neuroimaging data have advanced significantly since the beginning of neuroscience as a scientific discipline. Today, sophisticated statistical procedures allow us to examine complex multivariate patterns, however most of them are still constrained by assuming inherent linearity of neural processes. Here, we discuss a group of machine learning methods, called deep learning, which have drawn much attention in and outside the field of neuroscience in recent years and hold the potential to surpass the mentioned limitations. Firstly, we describe and explain the essential concepts in deep learning: the structure and the computational operations that allow deep models to learn. After that, we move to the most common applications of deep learning in neuroimaging data analysis: prediction of outcome, interpretation of internal representations, generation of synthetic data and segmentation. In the next section we present issues that deep learning poses, which concerns multidimensionality and multimodality of data, overfitting and computational cost, and propose possible solutions. Lastly, we discuss the current reach of DL usage in all the common applications in neuroimaging data analysis, where we consider the promise of multimodality, capability of processing raw data, and advanced visualization strategies. We identify research gaps, such as focusing on a limited number of criterion variables and the lack of a well-defined strategy for choosing architecture and hyperparameters. Furthermore, we talk about the possibility of conducting research with constructs that have been ignored so far or/and moving toward frameworks, such as RDoC, the potential of transfer learning and generation of synthetic data.

A transformer-Based neural language model that synthesizes brain activation maps from free-form text queries

Neuroimaging studies are often limited by the number of subjects and cognitive processes that can be feasibly interrogated. However, a rapidly growing number of neuroscientific studies have collectively accumulated an extensive wealth of results. Digesting this growing literature and obtaining novel insights remains to be a major challenge, since existing meta-analytic tools are constrained to keyword queries. In this paper, we present Text2Brain, an easy to use tool for synthesizing brain activation maps from open-ended text queries. Text2Brain was built on a transformer-based neural network language model and a coordinate-based meta-analysis of neuroimaging studies. Text2Brain combines a transformer-based text encoder and a 3D image generator, and was trained on variable-length text snippets and their corresponding activation maps sampled from 13,000 published studies. In our experiments, we demonstrate that Text2Brain can synthesize meaningful neural activation patterns from various free-form textual descriptions. Text2Brain is available at https://braininterpreter.com as a web-based tool for efficiently searching through the vast neuroimaging literature and generating new hypotheses.

Resting-state functional connectivity of social brain regions predicts motivated dishonesty

Motivated dishonesty is a typical social behavior varying from person to person. Resting-state fMRI (rsfMRI) is capable of identifying unique patterns from functional connectivity (FC) between brain regions. Recent work has built a link between brain networks in resting state to dishonesty in Western participants. To determine and reproduce the relevant neural patterns and build an interpretable model to predict dishonesty, we analyzed two conceptually similar datasets containing rsfMRI data with different dishonesty tasks. Both tasks implemented the information-passing paradigm, in which monetary rewards were employed to induce dishonesty. We applied connectome-based predictive modeling (CPM) to build a model among FC within and between four social brain networks (reward, self-referential, moral, and cognitive control). The CPM analysis indicated that FCs of social brain networks are predictive of dishonesty rate, especially FCs within reward network, and between self-referential and cognitive control networks. Our study offers an conceptual replication with integrated model to predict dishonesty with rsfMRI, and the results suggest that frequent motivated dishonest decisions may require the higher engagement of social brain regions.