Gender differences in cortical morphological networks

Cortical morphological networks for profiling autism spectrum disorder using tensor component analysis

Atypical neurodevelopmental disorders such as Autism Spectrum Disorder (ASD) can alter the cortex morphology at different levels: (i) a low-order level where cortical regions are examined individually, (ii) a high-order level where the relationship between two cortical regions is considered, and (iii) a multi-view high-order level where the relationship between regions is examined across multiple brain views. In this study, we propose to use the emerging multi-view cortical morphological network (CMN), which is derived from T1-w magnetic resonance imaging (MRI), to profile autistic and typical brains and pursue new ways of fingerprinting 'cortical morphology' at the intersection of 'network neuroscience'. Each CMN view models the pairwise morphological dissimilarity at the connection level using a specific cortical attribute (e.g., thickness). Specifically, we set out to identify the inherently most representative morphological connectivities shared across different views of the cortex in both autistic and normal control (NC) populations using tensor component analysis. We thus discover the connectional profiles of both populations shared across different CMNs of the left and right hemispheres, respectively. One of the most representative morphological cortical attributes for assessing the abnormal brain structures in patients with ASD is cortical thickness. The most representative morphological connectivities in multi-view CMN population of normal control and ASD subjects, respectively, and in both left and right hemispheres within the temporal, frontal, and insular lobes of individuals with ASD. These representative connectivities are corresponded to specific clinical features observed in individuals with ASD.

Age- and gender-related dispersion of brain networks across the lifespan

The effects of age and gender on large-scale resting-state networks (RSNs) reflecting within- and between-network connectivity in the healthy brain remain unclear. This study investigated how age and gender influence the brain network roles and topological properties underlying the ageing process. Ten RSNs were constructed based on 998 participants from the REST-meta-MDD cohort. Multivariate linear regression analysis was used to examine the independent and interactive influences of age and gender on large-scale RSNs and their topological properties. A support vector regression model integrating whole-brain network features was used to predict brain age across the lifespan and cognitive decline in an Alzheimer's disease spectrum (ADS) sample. Differential effects of age and gender on brain network roles were demonstrated across the lifespan. Specifically, cingulo-opercular, auditory, and visual (VIS) networks showed more incohesive features reflected by decreased intra-network connectivity with ageing. Further, females displayed distinctive brain network trajectory patterns in middle-early age, showing enhanced network connectivity within the fronto-parietal network (FPN) and salience network (SAN) and weakened network connectivity between the FPN-somatomotor, FPN-VIS, and SAN-VIS networks. Age - but not gender - induced widespread decrease in topological properties of brain networks. Importantly, these differential network features predicted brain age and cognitive impairment in the ADS sample. By showing that age and gender exert specific dispersion of dynamic network roles and trajectories across the lifespan, this study has expanded our understanding of age- and gender-related brain changes with ageing. Moreover, the findings may be useful for detecting early-stage dementia.

Comparative survey of multigraph integration methods for holistic brain connectivity mapping

One of the greatest scientific challenges in network neuroscience is to create a representative map of a population of heterogeneous brain networks, which acts as a connectional fingerprint. The connectional brain template (CBT), also named network atlas, presents a powerful tool for capturing the most representative and discriminative traits of a given population while preserving its topological patterns. The idea of a CBT is to integrate a population of heterogeneous brain connectivity networks, derived from different neuroimaging modalities or brain views (e.g., structural and functional), into a unified holistic representation. Here we review current state-of-the-art methods designed to estimate well-centered and representative CBT for populations of single-view and multi-view brain networks. We start by reviewing each CBT learning method, then we introduce the evaluation measures to compare CBT representativeness of populations generated by single-view and multigraph integration methods, separately, based on the following criteria: Centeredness, biomarker-reproducibility, node-level similarity, global-level similarity, and distance-based similarity. We demonstrate that the deep graph normalizer (DGN) method significantly outperforms other multi-graph and all single-view integration methods for estimating CBTs using a variety of healthy and disordered datasets in terms of centeredness, reproducibility (i.e., graph-derived biomarkers reproducibility that disentangle the typical from the atypical connectivity variability), and preserving the topological traits at both local and global graph-levels.

Predicting the evolution trajectory of population-driven connectional brain templates using recurrent multigraph neural networks

The mapping of the time-dependent evolution of the human brain connectivity using longitudinal and multimodal neuroimaging datasets provides insights into the development of neurological disorders and the way they alter the brain morphology, structure and function over time. Recently, the connectional brain template (CBT) was introduced as a compact representation integrating a population of brain multigraphs, where two brain regions can have multiple connections, into a single graph. Given a population of brain multigraphs observed at a baseline timepoint t₁, we aim to learn how to predict the evolution of the population CBT at follow-up timepoints t>t₁. Such model will allow us to foresee the evolution of the connectivity patterns of healthy and disordered individuals at the population level. Here we present recurrent multigraph integrator network (ReMI-Net^⋆) to forecast population templates at consecutive timepoints from a given single timepoint. In particular, we unprecedentedly design a graph neural network architecture to model the changes in the brain multigraph and identify the biomarkers that differentiate between the typical and atypical populations. Addressing such issues is of paramount importance in diagnosing neurodegenerative disorders at early stages and promoting new clinical studies based on the pinned-down biomarker brain regions or connectivities. In this paper, we demonstrate the design and use of the ReMI-Net^⋆ model, which learns both the multigraph node level and time level dependencies concurrently. Thanks to its novel graph convolutional design and normalization layers, ReMI-Net^⋆ predicts well-centered, discriminative, and topologically sound connectional templates over time. Additionally, the results show that our model outperforms all benchmarks and state-of-the-art methods by comparing and discovering the atypical connectivity alterations over time. Our ReMI-Net^⋆ code is available on GitHub at https://github.com/basiralab/ReMI-Net-Star.

Comprehensive Cortical Structural Features Predict the Efficacy of Cognitive Behavioral Therapy in Obsessive-Compulsive Disorder

Although cognitive behavioral therapy (CBT) is effective for patients with obsessive-compulsive disorder (OCD), 40% of OCD patients show a poor response to CBT. This study aimed to identify the cortical structural factors that predict CBT outcomes in OCD patients. A total of 56 patients with OCD received baseline structural MRI (sMRI) scanning and 14 individual CBT sessions. The linear support vector regression (SVR) models were used to identify the predictive performance of sMRI indices, including gray matter volume, cortical thickness, sulcal depth, and gyrification value. The patients’ OC symptoms decreased significantly after CBT intervention (p < 0.001). We found the model with the comprehensive variables exhibited better performance than the models with single structural indices (MAE = 0.14, MSE = 0.03, R2 = 0.36), showing a significant correlation between the true value and the predicted value (r = 0.63, p < 0.001). The results indicated that a model integrating four cortical structural features can accurately predict the effectiveness of CBT for OCD. Future models incorporating other brain indicators, including brain functional indicators, EEG indicators, neurotransmitters, etc., which might be more accurate for predicting the effectiveness of CBT for OCD, are needed.

Multigraph classification using learnable integration network with application to gender fingerprinting

Multigraphs with heterogeneous views present one of the most challenging obstacles to classification tasks due to their complexity. Several works based on feature selection have been recently proposed to disentangle the problem of multigraph heterogeneity. However, such techniques have major drawbacks. First, the bulk of such works lies in the vectorization and the flattening operations, failing to preserve and exploit the rich topological properties of the multigraph. Second, they learn the classification process in a dichotomized manner where the cascaded learning steps are pieced in together independently. Hence, such architectures are inherently agnostic to the cumulative estimation error from step to step. To overcome these drawbacks, we introduce MICNet (multigraph integration and classifier network), the first end-to-end graph neural network based model for multigraph classification. First, we learn a single-view graph representation of a heterogeneous multigraph using a GNN based integration model. The integration process in our model helps tease apart the heterogeneity across the different views of the multigraph by generating a subject-specific graph template while preserving its geometrical and topological properties conserving the node-wise information while reducing the size of the graph (i.e., number of views). Second, we classify each integrated template using a geometric deep learning block which enables us to grasp the salient graph features. We train, in end-to-end fashion, these two blocks using a single objective function to optimize the classification performance. We evaluate our MICNet in gender classification using brain multigraphs derived from different cortical measures. We demonstrate that our MICNet significantly outperformed its variants thereby showing its great potential in multigraph classification.

Classification of multiple sclerosis patients based on structural disconnection: A robust feature selection approach

Background and Purpose

Although structural disconnection represents the hallmark of multiple sclerosis (MS) pathophysiology, classification attempts based on structural connectivity have achieved low accuracy levels. Here, we set out to fill this gap, exploring the performance of supervised classifiers on features derived from microstructure informed tractography and selected applying a novel robust approach.

Methods

Using microstructure informed tractography with diffusion MRI data, we created quantitative connectomes of 55 MS patients and 24 healthy controls. We then used a robust approach—based on two classical methods of feature selection— to select relevant features from three network representations (whole connectivity matrices, node strength, and local efficiency). Classification accuracy of the selected features was tested with five different classifiers, while their meaningfulness was tested via correlation with clinical scales. As a comparison, the same classifiers were run on features selected with the standard procedure in network analysis (thresholding).

Results

Our procedure identified 11 features for the whole net, five for local efficiency, and seven for node strength. For all classifiers, the accuracy was in the range 64.5%‐91.1%, with features extracted from the whole net reaching the maximum, and overcoming results obtained with the standard procedure in all cases. Correlations with clinical scales were identified across functional domains, from motor and cognitive abilities to fatigue and depression.

Conclusion

Applying a robust feature selection procedure to quantitative structural connectomes, we were able to classify MS patients with excellent accuracy, while providing information on the white matter connections and gray matter regions more affected by MS pathology.

A comparative study of machine learning methods for predicting the evolution of brain connectivity from a baseline timepoint

BACKGROUND

Predicting the evolution of the brain network, also called connectome, by foreseeing changes in the connectivity weights linking pairs of anatomical regions makes it possible to spot connectivity-related neurological disorders in earlier stages and detect the development of potential connectomic anomalies. Remarkably, such a challenging prediction problem remains least explored in the predictive connectomics literature. It is a known fact that machine learning (ML) methods have proven their predictive abilities in a wide variety of computer vision problems. However, ML techniques specifically tailored for the prediction of brain connectivity evolution trajectory from a single timepoint are almost absent.

NEW METHOD

To fill this gap, we organized a Kaggle competition where 20 competing teams designed advanced machine learning pipelines for predicting the brain connectivity evolution from a single timepoint. The teams developed their ML pipelines with combination of data pre-processing, dimensionality reduction and learning methods. Each ML framework inputs a baseline brain connectivity matrix observed at baseline timepoint t₀ and outputs the brain connectivity map at a follow-up timepoint t₁. The longitudinal OASIS-2 dataset was used for model training and evaluation. Both random data split and 5-fold cross-validation strategies were used for ranking and evaluating the generalizability and scalability of each competing ML pipeline.

RESULTS

Utilizing an inclusive approach, we ranked the methods based on two complementary evaluation metrics (mean absolute error (MAE) and Pearson Correlation Coefficient (PCC)) and their performances using different training and testing data perturbation strategies (single random split and cross-validation). The final rank was calculated using the rank product for each competing team across all evaluation measures and validation strategies. Furthermore, we added statistical significance values to each proposed pipeline.

CONCLUSION

In support of open science, the developed 20 ML pipelines along with the connectomic dataset are made available on GitHub (https://github.com/basiralab/Kaggle-BrainNetPrediction-Toolbox). The outcomes of this competition are anticipated to lead the further development of predictive models that can foresee the evolution of the brain connectivity over time, as well as other types of networks (e.g., genetic networks).

Kinesthetic and visual imagery in young adults with chronic neck pain

Background:Young adults complain of neck pain almost every year. In recent years the ability of motor imagery (kinesthetic and visual imagery) in many musculoskeletal system problems other than neck pain in young adults has been investigated in the literature. The Cross-Sectional study aimed to question motor imagery ability in young adults with chronic neck pain. Methods: Two groups were included in the study: the chronic neck pain group (n = 83) and the control group (n = 91). Motor imagery ability of both groups was evaluated with Movement Imagery Questionnaire-3. Additionally, in the chronic neck pain group, pain was evaluated with the Short Form-McGill Pain Questionnaire, disability was evaluated with the Neck Disability Index, and kinesiophobia was evaluated with Tampa Scale for Kinesiophobia. Conclusions: Internal visual imagery and kinesthetic imagery were significantly different between chronic neck pain and control groups. There was a negative linear relationship between disability and internal visual imagery, external visual imagery, and kinesthetic imagery. Motor imagery ability is reduced in young adults with chronic neck pain. In addition, as the severity of disability increases, the motor imagery ability decreases. Therefore, it is considered appropriate to include a motor imagery training program when treating chronic neck pain in the future.

Sex-specific signatures of intrinsic hippocampal networks and regional integrity underlying cognitive status in multiple sclerosis

The hippocampus is an anatomically compartmentalized structure embedded in highly wired networks that are essential for cognitive functions. The hippocampal vulnerability has been postulated in acute and chronic neuroinflammation in multiple sclerosis, while the patterns of occurring inflammation, neurodegeneration or compensation have not yet been described. Besides focal damage to hippocampal tissue, network disruption is an important contributor to cognitive decline in multiple sclerosis patients. We postulate sex-specific trajectories in hippocampal network reorganization and regional integrity and address their relationship to markers of neuroinflammation, cognitive/memory performance and clinical severity. In a large cohort of multiple sclerosis patients (n = 476; 337 females, age 35 ± 10 years, disease duration 16 ± 14 months) and healthy subjects (n = 110, 54 females; age 34 ± 15 years), we utilized MRI at baseline and at 2-year follow-up to quantify regional hippocampal volumetry and reconstruct single-subject hippocampal networks. Through graph analytical tools we assessed the clustered topology of the hippocampal networks. Mixed-effects analyses served to model sex-based differences in hippocampal network and subfield integrity between multiple sclerosis patients and healthy subjects at both time points and longitudinally. Afterwards, hippocampal network and subfield integrity were related to clinical and radiological variables in dependency of sex attribution. We found a more clustered network architecture in both female and male patients compared to their healthy counterparts. At both time points, female patients displayed a more clustered network topology in comparison to male patients. Over time, multiple sclerosis patients developed an even more clustered network architecture, though with a greater magnitude in females. We detected reduced regional volumes in most of the addressed hippocampal subfields in both female and male patients compared to healthy subjects. Compared to male patients, females displayed lower volumes of para- and presubiculum but higher volumes of the molecular layer. Longitudinally, volumetric alterations were more pronounced in female patients, which showed a more extensive regional tissue loss. Despite a comparable cognitive/memory performance between female and male patients over the follow-up period, we identified a strong interrelation between hippocampal network properties and cognitive/memory performance only in female patients. Our findings evidence a more clustered hippocampal network topology in female patients with a more extensive subfield volume loss over time. A stronger relation between cognitive/memory performance and the network topology in female patients suggests greater entrainment of the brain’s reserve. These results may serve to adapt sex-targeted neuropsychological interventions.

The development of structural covariance networks during the transition from childhood to adolescence

Structural covariance conceptualizes how morphologic properties of brain regions are related to one another (across individuals). It can provide unique information to cortical structure (e.g., thickness) about the development of functionally meaningful networks. The current study investigated how structural covariance networks develop during the transition from childhood to adolescence, a period characterized by marked structural re-organization. Participants (N = 192; scans = 366) completed MRI assessments between 8.5 and 14.5 years of age. A sliding window approach was used to create “age-bins”, and structural covariance networks (based on cortical thickness) were created for each bin. Next, generalized additive models were used to characterize trajectories of age-related changes in network properties. Results revealed nonlinear trajectories with “peaks” in mean correlation and global density that are suggestive of a period of convergence in anatomical properties across the cortex during early adolescence, prior to regional specialization. “Hub” regions in sensorimotor cortices were present by late childhood, but the extent and strength of association cortices as “hubs” increased into mid-adolescence. Moreover, these regional changes were found to be related to rates of thinning across the cortex. In the context of neurocognitive networks, the frontoparietal, default mode, and attention systems exhibited age-related increases in within-network and between-network covariance. These regional and modular developmental patterns are consistent with continued refinement of socioemotional and other complex executive functions that are supported by higher-order cognitive networks during early adolescence.

Sex effects on cortical morphological networks in healthy young adults

Understanding sex-related differences across the human cerebral cortex is an important step in elucidating the basis of psychological, behavioural and clinical differences between the sexes. Prior structural neuroimaging studies primarily focused on regional sex differences using univariate analyses. Here we focus on sex differences in cortical morphological networks (CMNs) derived using multivariate modelling of regional cortical measures of volume and surface from high-quality structural MRI scans from healthy participants in the Human Connectome Project (HCP) (n = 1,063) and the Southwest University Longitudinal Imaging Multimodal (SLIM) study (n = 549). The functional relevance of the CMNs was inferred using the NeuroSynth decoding function. Sex differences were widespread but not uniform. In general, females had higher volume, thickness and cortical folding in networks that involve prefrontal (both ventral and dorsal regions including the anterior cingulate) and parietal regions while males had higher volume, thickness and cortical folding in networks that primarily include temporal and posterior cortical regions. CMN loading coefficients were used as input features to linear discriminant analyses that were performed separately in the HCP and SLIM; sex was predicted with a high degree of accuracy (81%-85%) across datasets. The availability of behavioral data in the HCP enabled us to show that male-biased surface-based CMNs were associated with externalizing behaviors. These results extend previous literature on regional sex-differences by identifying CMNs that can reliably predict sex, are relevant to the expression of psychopathology and provide the foundation for the future investigation of their functional significance in clinical populations.

Brain graph synthesis by dual adversarial domain alignment and target graph prediction from a source graph

Developing predictive intelligence in neuroscience for learning how to generate multimodal medical data from a single modality can improve neurological disorder diagnosis with minimal data acquisition resources. Existing deep learning frameworks are mainly tailored for images, which might fail in handling geometric data (e.g., brain graphs). Specifically, predicting a target brain graph from a single source brain graph remains largely unexplored. Solving such problem is generally challenged with domain fracturecaused by the difference in distribution between source and target domains. Besides, solving the prediction and domain fracture independently might not be optimal for both tasks. To address these challenges, we unprecedentedly propose a Learning-guided Graph Dual Adversarial Domain Alignment (LG-DADA) framework for predicting a target brain graph from a source brain graph. The proposed LG-DADA is grounded in three fundamental contributions: (1) a source data pre-clustering step using manifold learning to firstly handle source data heterogeneity and secondly circumvent mode collapse in generative adversarial learning, (2) a domain alignment of source domain to the target domain by adversarially learning their latent representations, and (3) a dual adversarial regularization that jointly learns a source embedding of training and testing brain graphs using two discriminators and predict the training target graphs. Results on morphological brain graphs synthesis showed that our method produces better prediction accuracy and visual quality as compared to other graph synthesis methods.

Estimation of gender-specific connectional brain templates using joint multi-view cortical morphological network integration

The estimation of a connectional brain template (CBT) integrating a population of brain networks while capturing shared and differential connectional patterns across individuals remains unexplored in gender fingerprinting. This paper presents the first study to estimate gender-specific CBTs using multi-view cortical morphological networks (CMNs) estimated from conventional T1-weighted magnetic resonance imaging (MRI). Specifically, each CMN view is derived from a specific cortical attribute (e.g. thickness), encoded in a network quantifying the dissimilarity in morphology between pairs of cortical brain regions. To this aim, we propose Multi-View Clustering and Fusion Network (MVCF-Net), a novel multi-view network fusion method, which can jointly identify consistent and differential clusters of multi-view datasets in order to capture simultaneously similar and distinct connectional traits of samples. Our MVCF-Net method estimates a representative and well-centered CBTs for male and female populations, independently, to eventually identify their fingerprinting regions of interest (ROIs) in four main steps. First, we perform multi-view network clustering model based on manifold optimization which groups CMNs into shared and differential clusters while preserving their alignment across views. Second, for each view, we linearly fuse CMNs belonging to each cluster, producing local CBTs. Third, for each cluster, we non-linearly integrate the local CBTs across views, producing a cluster-specific CBT. Finally, by linearly fusing the cluster-specific centers we estimate a final CBT of the input population. MVCF-Net produced the most centered and representative CBTs for male and female populations and identified the most discriminative ROIs marking gender differences. The most two gender-discriminative ROIs involved the lateral occipital cortex and pars opercularis in the left hemisphere and the middle temporal gyrus and lingual gyrus in the right hemisphere.