Probing large language models (LLMs) has yielded valuable insights into their internal mechanisms by linking neural representations to interpretable semantics. However, how neurons functionally co-activate with each other to give rise to emergent capabilities remains largely unknown, hindering a deeper understanding and safer development of LLMs. In this work, we introduce graph probing, a method for uncovering the functional connectivity topology of LLM neurons and relating it to language generation performance. By analyzing internal neural graphs across diverse LLM families and scales, we discover a universal predictability of next-token prediction performance using only neural topology. This predictability is robust even when retaining just 1% of neuron connections or probing models after only 8 pretraining steps, highlighting the sparsity and early emergence of topological patterns. Further graph matching analysis suggests that, despite significant distinctions in architectures, parameters, and training data, different LLMs develop intricate and consistent neural topological structures that may form the foundation for their language generation abilities. Codes and data for the graph probing toolbox are released at https://github.com/DavyMorgan/llm-graph-probing.
Neuropathic pain, affecting up to 10% of adults, remains difficult to treat due to limited therapeutic efficacy and tolerability. Although resting-state functional MRI (rs-fMRI) is a promising non-invasive measurement of brain biomarkers to predict drug response in therapeutic development, the complexity of fMRI demands machine learning models with substantial capacity. However, extreme data scarcity in neuropathic pain research limits the application of high-capacity models. To address the challenge of data scarcity, we propose FMM$_{TC}$, a Foundation-Model-boosted Multimodal learning framework for fMRI-based neuropathic pain drug response prediction, which leverages both internal multimodal information in pain-specific data and external knowledge from large pain-agnostic data. Specifically, to maximize the value of limited pain-specific data, FMM$_{TC}$ integrates complementary information from two rs-fMRI modalities: Time series and functional Connectivity. FMM$_{TC}$ is further boosted by an fMRI foundation model with its external knowledge from extensive pain-agnostic fMRI datasets enriching limited pain-specific information. Evaluations with an in-house dataset and a public dataset from OpenNeuro demonstrate FMM$_{TC}$'s superior representation ability, generalizability, and cross-dataset adaptability over existing unimodal fMRI models that only consider one of the rs-fMRI modalities. The ablation study validates the effectiveness of multimodal learning and foundation-model-powered external knowledge transfer in FMM$_{TC}$. An integrated gradient-based interpretation study explains how FMM$_{TC}$'s cross-dataset dynamic behaviors enhance its adaptability. In conclusion, FMM$_{TC}$ boosts clinical trials in neuropathic pain therapeutic development by accurately predicting drug responses to improve the participant stratification efficiency.
With recent advancements in non-invasive techniques for measuring brain activity, such as magnetic resonance imaging (MRI), the study of structural and functional brain networks through graph signal processing (GSP) has gained notable prominence. GSP stands as a key tool in unraveling the interplay between the brain's function and structure, enabling the analysis of graphs defined by the connections between regions of interest -- referred to as connectomes in this context. Our work represents a further step in this direction by exploring supervised contrastive learning methods within the realm of graph representation learning. The main objective of this approach is to generate subject-level (i.e., graph-level) vector representations that bring together subjects sharing the same label while separating those with different labels. These connectome embeddings are derived from a graph neural network Encoder-Decoder architecture, which jointly considers structural and functional connectivity. By leveraging data augmentation techniques, the proposed framework achieves state-of-the-art performance in a gender classification task using Human Connectome Project data. More broadly, our connectome-centric methodological advances support the promising prospect of using GSP to discover more about brain function, with potential impact to understanding heterogeneity in the neurodegeneration for precision medicine and diagnosis.
Despite advances in deep learning for estimating brain age from structural MRI data, incorporating functional MRI data is challenging due to its complex structure and the noisy nature of functional connectivity measurements. To address this, we present the Multitask Adversarial Variational Autoencoder, a custom deep learning framework designed to improve brain age predictions through multimodal MRI data integration. This model separates latent variables into generic and unique codes, isolating shared and modality-specific features. By integrating multitask learning with sex classification as an additional task, the model captures sex-specific aging patterns. Evaluated on the OpenBHB dataset, a large multisite brain MRI collection, the model achieves a mean absolute error of 2.77 years, outperforming traditional methods. This success positions M-AVAE as a powerful tool for metaverse-based healthcare applications in brain age estimation.
Functional brain network properties are heavily influenced by how the the network nodes are defined. A common approach uses Regions of Interest (ROIs), i.e., predetermined collections of functional magnetic resonance imaging (fMRI) measurement voxels, as nodes. Their definition is always a compromise, as static ROIs cannot capture the dynamics and temporal reconfigurations of the brain areas. Consequently, the ROIs do not align with the functionally homogeneous regions, which can explain the low functional homogeneity values observed for the ROIs. This is in violation of the underlying homogeneity assumption in functional brain network analysis pipelines, which can cause serious problems such as spurious network structure. We introduce the node-reconfiguring multilayer network model, where nodes represent ROIs with boundaries optimized for high functional homogeneity in each time window. In this representation, network layers correspond to time windows, intralayer links depict functional connectivity between ROIs, and interlayer links quantify the overlap between ROIs on different layers. The ROI optimization approach increases functional homogeneity notably, yielding an over 10-fold increase in the fraction of ROIs with high homogeneity compared to static ROIs from the Brainnetome atlas. The optimized ROIs reorganize non-trivially at short time scales of consecutive time windows and across several windows. The amount of reorganization across time windows is connected to intralayer hubness: ROIs with intermediate levels of reorganization have stronger intralayer links than extremely stable or unstable ROIs. Our results demonstrate that reconfiguring parcellations yield more accurate network models of brain function. This supports the ongoing paradigm shift towards the chronnectome that sees the brain as a set of sources with continuously reconfiguring spatial and connectivity profiles.
Although modern imaging technologies allow us to study connectivity between two distinct brain regions in-vivo, an in-depth understanding of how anatomical structure supports brain function and how spontaneous functional fluctuations emerge remarkable cognition is still elusive. Meanwhile, tremendous efforts have been made in the realm of machine learning to establish the nonlinear mapping between neuroimaging data and phenotypic traits. However, the absence of neuroscience insight in the current approaches poses significant challenges in understanding cognitive behavior from transient neural activities. To address this challenge, we put the spotlight on the coupling mechanism of structural connectivity (SC) and functional connectivity (FC) by formulating such network neuroscience question into an expressive graph representation learning problem for high-order topology. Specifically, we introduce the concept of topological detour to characterize how a ubiquitous instance of FC (direct link) is supported by neural pathways (detour) physically wired by SC, which forms a cyclic loop interacted by brain structure and function. In the clich\'e of machine learning, the multi-hop detour pathway underlying SC-FC coupling allows us to devise a novel multi-head self-attention mechanism within Transformer to capture multi-modal feature representation from paired graphs of SC and FC. Taken together, we propose a biological-inspired deep model, coined as NeuroPath, to find putative connectomic feature representations from the unprecedented amount of neuroimages, which can be plugged into various downstream applications such as task recognition and disease diagnosis. We have evaluated NeuroPath on large-scale public datasets including HCP and UK Biobank under supervised and zero-shot learning, where the state-of-the-art performance by our NeuroPath indicates great potential in network neuroscience.
Autism Spectrum Disorder (ASD) is a neurodevelopmental condition that encompasses a wide variety of symptoms and degrees of impairment, which makes the diagnosis and treatment challenging. Functional magnetic resonance imaging (fMRI) has been extensively used to study brain activity in ASD, and machine learning methods have been applied to analyze resting state fMRI (rs-fMRI) data. However, fewer studies have explored the recent transformer-based models on rs-fMRI data. Given the superiority of transformer models in capturing long-range dependencies in sequence data, we have developed a transformer-based self-supervised framework that directly analyzes time-series fMRI data without computing functional connectivity. To address over-fitting in small datasets and enhance the model performance, we propose self-supervised pre-training tasks to reconstruct the randomly masked fMRI time-series data, investigating the effects of various masking strategies. We then finetune the model for the ASD classification task and evaluate it using two public datasets and five-fold cross-validation with different amounts of training data. The experiments show that randomly masking entire ROIs gives better model performance than randomly masking time points in the pre-training step, resulting in an average improvement of 10.8% for AUC and 9.3% for subject accuracy compared with the transformer model trained from scratch across different levels of training data availability. Our code is available on GitHub.
The aim of this paper is threefold. We inform the AI practitioner about the human visual system with an extensive literature review; we propose a novel biologically motivated neural network for image classification; and, finally, we present a new plug-and-play module to model context awareness. We focus on the effect of incorporating circuit motifs found in biological brains to address visual recognition. Our convolutional architecture is inspired by the connectivity of human cortical and subcortical streams, and we implement bottom-up and top-down modulations that mimic the extensive afferent and efferent connections between visual and cognitive areas. Our Contextual Attention Block is simple and effective and can be integrated with any feed-forward neural network. It infers weights that multiply the feature maps according to their causal influence on the scene, modeling the co-occurrence of different objects in the image. We place our module at different bottlenecks to infuse a hierarchical context awareness into the model. We validated our proposals through image classification experiments on benchmark data and found a consistent improvement in performance and the robustness of the produced explanations via class activation. Our code is available at https://github.com/gianlucarloni/CoCoReco.
We propose a novel approach to investigate the brain mechanisms that support coordination of behavior between individuals. Brain states in single individuals defined by the patterns of functional connectivity between brain regions are used to create joint symbolic representations of the evolution of brain states in two or more individuals performing a task together. These symbolic dynamics can be analyzed to reveal aspects of the dynamics of joint brain states that are related to coordination or other interactive behaviors. We apply this approach to simultaneous electroencephalographic (EEG) data from pairs of subjects engaged in two different modes of finger-tapping coordination tasks (synchronization and syncopation) under different interaction conditions (Uncoupled, Leader-Follower, and Mutual) to explore the neural mechanisms of multi-person motor coordination. Our results reveal that the dyads exhibit mostly the same joint symbols in different interaction conditions - the most important differences are reflected in the symbolic dynamics. Recurrence analysis shows that interaction influences the dwell time in specific joint symbols and the structure of joint symbol sequences (motif length). In synchronization, increasing feedback promotes stability with longer dwell times and motif length. In syncopation, Leader-Follower interactions enhance stability (increase dwell time and motif length), but Mutual feedback dramatically reduces stability. Network analysis reveals distinct topological changes with task and feedback. In synchronization, stronger coupling stabilizes a few states restricting the pattern of flow between states, preserving a core-periphery structure of the joint brain states. In syncopation, a more distributed flow amongst a larger set of joint brain states reduces the dominance of core joint brain states.
The segregation and integration of infant brain networks undergo tremendous changes due to the rapid development of brain function and organization. Traditional methods for estimating brain modularity usually rely on group-averaged functional connectivity (FC), often overlooking individual variability. To address this, we introduce a novel approach utilizing Bayesian modeling to analyze the dynamic development of functional modules in infants over time. This method retains inter-individual variability and, in comparison to conventional group averaging techniques, more effectively detects modules, taking into account the stationarity of module evolution. Furthermore, we explore gender differences in module development under awake and sleep conditions by assessing modular similarities. Our results show that female infants demonstrate more distinct modular structures between these two conditions, possibly implying relative quiet and restful sleep compared with male infants.
Brain-computer interfaces (BCIs) enable users to interact with the external world using brain activity. Despite their potential in neuroscience and industry, BCI performance remains inconsistent in noninvasive applications, often prioritizing algorithms that achieve high classification accuracies while masking the neural mechanisms driving that performance. In this study, we investigated the interpretability of features derived from brain network lateralization, benchmarking against widely used techniques like power spectrum density (PSD), common spatial pattern (CSP), and Riemannian geometry. We focused on the spatial distribution of the functional connectivity within and between hemispheres during motor imagery tasks, introducing network-based metrics such as integration and segregation. Evaluating these metrics across multiple EEG-based BCI datasets, our findings reveal that network lateralization offers neurophysiological plausible insights, characterized by stronger lateralization in sensorimotor and frontal areas contralateral to imagined movements. While these lateralization features did not outperform CSP and Riemannian geometry in terms of classification accuracy, they demonstrated competitive performance against PSD alone and provided biologically relevant interpretation. This study underscores the potential of brain network lateralization as a new feature to be integrated in motor imagery-based BCIs for enhancing the interpretability of noninvasive applications.
Background: Deep learning models have shown promise in diagnosing neurodevelopmental disorders (NDD) like ASD and ADHD. However, many models either use graph neural networks (GNN) to construct single-level brain functional networks (BFNs) or employ spatial convolution filtering for local information extraction from rs-fMRI data, often neglecting high-order features crucial for NDD classification. Methods: We introduce a Multi-view High-order Network (MHNet) to capture hierarchical and high-order features from multi-view BFNs derived from rs-fMRI data for NDD prediction. MHNet has two branches: the Euclidean Space Features Extraction (ESFE) module and the Non-Euclidean Space Features Extraction (Non-ESFE) module, followed by a Feature Fusion-based Classification (FFC) module for NDD identification. ESFE includes a Functional Connectivity Generation (FCG) module and a High-order Convolutional Neural Network (HCNN) module to extract local and high-order features from BFNs in Euclidean space. Non-ESFE comprises a Generic Internet-like Brain Hierarchical Network Generation (G-IBHN-G) module and a High-order Graph Neural Network (HGNN) module to capture topological and high-order features in non-Euclidean space. Results: Experiments on three public datasets show that MHNet outperforms state-of-the-art methods using both AAL1 and Brainnetome Atlas templates. Extensive ablation studies confirm the superiority of MHNet and the effectiveness of using multi-view fMRI information and high-order features. Our study also offers atlas options for constructing more sophisticated hierarchical networks and explains the association between key brain regions and NDD. Conclusion: MHNet leverages multi-view feature learning from both Euclidean and non-Euclidean spaces, incorporating high-order information from BFNs to enhance NDD classification performance.
Objective: fMRI and derived measures such as functional connectivity (FC) have been used to predict brain age, general fluid intelligence, psychiatric disease status, and preclinical neurodegenerative disease. However, it is not always clear that all demographic confounds, such as age, sex, and race, have been removed from fMRI data. Additionally, many fMRI datasets are restricted to authorized researchers, making dissemination of these valuable data sources challenging. Methods: We create a variational autoencoder (VAE)-based model, DemoVAE, to decorrelate fMRI features from demographics and generate high-quality synthetic fMRI data based on user-supplied demographics. We train and validate our model using two large, widely used datasets, the Philadelphia Neurodevelopmental Cohort (PNC) and Bipolar and Schizophrenia Network for Intermediate Phenotypes (BSNIP). Results: We find that DemoVAE recapitulates group differences in fMRI data while capturing the full breadth of individual variations. Significantly, we also find that most clinical and computerized battery fields that are correlated with fMRI data are not correlated with DemoVAE latents. An exception are several fields related to schizophrenia medication and symptom severity. Conclusion: Our model generates fMRI data that captures the full distribution of FC better than traditional VAE or GAN models. We also find that most prediction using fMRI data is dependent on correlation with, and prediction of, demographics. Significance: Our DemoVAE model allows for generation of high quality synthetic data conditioned on subject demographics as well as the removal of the confounding effects of demographics. We identify that FC-based prediction tasks are highly influenced by demographic confounds.
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.
To date, most discoveries of network subcomponents that implement human-interpretable computations in deep vision models have involved close study of single units and large amounts of human labor. We explore scalable methods for extracting the subgraph of a vision model's computational graph that underlies recognition of a specific visual concept. We introduce a new method for identifying these subgraphs: specifying a visual concept using a few examples, and then tracing the interdependence of neuron activations across layers, or their functional connectivity. We find that our approach extracts circuits that causally affect model output, and that editing these circuits can defend large pretrained models from adversarial attacks.
Diagnosis of mild cognitive impairment (MCI) and subjective cognitive decline (SCD) from fMRI functional connectivity (FC) has gained popularity, but most FC-based diagnostic models are black boxes lacking casual reasoning so they contribute little to the knowledge about FC-based neural biomarkers of cognitive decline.To enhance the explainability of diagnostic models, we propose a generative counterfactual attention-guided network (GCAN), which introduces counterfactual reasoning to recognize cognitive decline-related brain regions and then uses these regions as attention maps to boost the prediction performance of diagnostic models. Furthermore, to tackle the difficulty in the generation of highly-structured and brain-atlas-constrained FC, which is essential in counterfactual reasoning, an Atlas-Aware Bidirectional Transformer (AABT) method is developed. AABT employs a bidirectional strategy to encode and decode the tokens from each network of brain atlas, thereby enhancing the generation of high-quality target label FC. In the experiments of hospital-collected and ADNI datasets, the generated attention maps closely resemble FC abnormalities in the literature on SCD and MCI. The diagnostic performance is also superior to baseline models. The code is available at https://github.com/SXR3015/GCAN
Researchers continue exploring neurons' intricate patterns of activity in the cerebral visual cortex in response to visual stimuli. The way neurons communicate and optimize their interactions with each other under different experimental conditions remains a topic of active investigation. Probabilistic Graphical Models are invaluable tools in neuroscience research, as they let us identify the functional connections, or conditional statistical dependencies, between neurons. Graphical models represent these connections as a graph, where nodes represent neurons and edges indicate the presence of functional connections between them. We developed the R package neurofuncon for the computation and visualization of functional connectivity graphs from large-scale data based on the Graphical lasso. We illustrate the use of this package with publicly available two-photon calcium microscopy imaging data from approximately 10000 neurons in a 1mm cubic section of a mouse visual cortex.
This paper presents a comprehensive and quality collection of functional human brain network data for potential research in the intersection of neuroscience, machine learning, and graph analytics. Anatomical and functional MRI images have been used to understand the functional connectivity of the human brain and are particularly important in identifying underlying neurodegenerative conditions such as Alzheimer's, Parkinson's, and Autism. Recently, the study of the brain in the form of brain networks using machine learning and graph analytics has become increasingly popular, especially to predict the early onset of these conditions. A brain network, represented as a graph, retains rich structural and positional information that traditional examination methods are unable to capture. However, the lack of publicly accessible brain network data prevents researchers from data-driven explorations. One of the main difficulties lies in the complicated domain-specific preprocessing steps and the exhaustive computation required to convert the data from MRI images into brain networks. We bridge this gap by collecting a large amount of MRI images from public databases and a private source, working with domain experts to make sensible design choices, and preprocessing the MRI images to produce a collection of brain network datasets. The datasets originate from 6 different sources, cover 4 brain conditions, and consist of a total of 2,702 subjects. We test our graph datasets on 12 machine learning models to provide baselines and validate the data quality on a recent graph analysis model. To lower the barrier to entry and promote the research in this interdisciplinary field, we release our brain network data and complete preprocessing details including codes at https://doi.org/10.17608/k6.auckland.21397377 and https://github.com/brainnetuoa/data_driven_network_neuroscience.
Resting-state fMRI (rs-fMRI) functional connectivity (FC) analysis provides valuable insights into the relationships between different brain regions and their potential implications for neurological or psychiatric disorders. However, specific design efforts to predict treatment response from rs-fMRI remain limited due to difficulties in understanding the current brain state and the underlying mechanisms driving the observed patterns, which limited the clinical application of rs-fMRI. To overcome that, we propose a graph learning framework that captures comprehensive features by integrating both correlation and distance-based similarity measures under a contrastive loss. This approach results in a more expressive framework that captures brain dynamic features at different scales and enables more accurate prediction of treatment response. Our experiments on the chronic pain and depersonalization disorder datasets demonstrate that our proposed method outperforms current methods in different scenarios. To the best of our knowledge, we are the first to explore the integration of distance-based and correlation-based neural similarity into graph learning for treatment response prediction.
The bulk of the research effort on brain connectivity revolves around statistical associations among brain regions, which do not directly relate to the causal mechanisms governing brain dynamics. Here we propose the multiscale causal backbone (MCB) of brain dynamics, shared by a set of individuals across multiple temporal scales, and devise a principled methodology to extract it. Our approach leverages recent advances in multiscale causal structure learning and optimizes the trade-off between the model fit and its complexity. Empirical assessment on synthetic data shows the superiority of our methodology over a baseline based on canonical functional connectivity networks. When applied to resting-state fMRI data, we find sparse MCBs for both the left and right brain hemispheres. Thanks to its multiscale nature, our approach shows that at low-frequency bands, causal dynamics are driven by brain regions associated with high-level cognitive functions; at higher frequencies instead, nodes related to sensory processing play a crucial role. Finally, our analysis of individual multiscale causal structures confirms the existence of a causal fingerprint of brain connectivity, thus supporting the existing extensive research in brain connectivity fingerprinting from a causal perspective.