A GICA-TVGL framework to study sex differences in resting state fMRI dynamic connectivity

Beyond network connectivity: A classification approach to brain age prediction with resting-state fMRI

The brain is a complex, dynamic organ that shows differences in the same subject at various periods. Understanding how brain activity changes across age as a function of the brain networks has been greatly abetted by fMRI. Canonical analysis consists of determining how alterations in connectivity patterns (CPs) of certain regions are affected. An alternative approach is taken here by not considering connectivity but rather features computed from recordings at the regions of interest (ROIs). Using machine learning (ML) we assess how neural signals are altered by and prospectively predictive of age and sex via a methodology that is novel in drawing upon pairwise classification across six decades of subjects' chronological ages. ML is used to answer the equally important questions of what properties of the computed features are most predictive as well as which brain networks are most affected by aging. It was found that there is decreased differentiation among the neural signals of older subjects that are separated in age by the same number of years as younger subjects. Furthermore, the burstiness of the signals change at different rates between males and females. The findings provide insight into brain aging via an ROI-based analysis, the consideration of several feature groups, and a novel classification-based ML pipeline. There is also a contribution to understanding the effects of data aggregated from different recording centers on the conclusions of fMRI studies.

Subsystem mechanisms of default mode network underlying white matter hyperintensity-related cognitive impairment

Functional changes of default mode network (DMN) have been proven to be closely associated with white matter hyperintensity (WMH) related cognitive impairment (CI). However, subsystem mechanisms of DMN underlying WMH-related CI remain unclear. The present study recruited WMH patients (n = 206) with mild CI and normal cognition, as well as healthy controls (HC, n = 102). Static/dynamic functional connectivity (FC) of the DMN's three subsystems were calculated using resting-state functional MRI. K-means clustering analyses were performed to extract distinct dynamic connectivity states. Compared with the WMH-NC group, the WMH-MCI group displayed lower static FC within medial temporal lobe (MTL) and core subsystem, between core-MTL subsystem, as well as between core and dorsal medial prefrontal cortex subsystem. All these static alterations were positively associated with information processing speed (IPS). Regarding dynamic FC, the WMH-MCI group exhibited higher dynamic FC within MTL subsystem than the HC and WMH-NC groups. Altered dynamic FC within MTL subsystem mediated the relationship between WMH and memory span (indirect effect: -0.2251, 95% confidence interval [-0.6295, -0.0267]). Additionally, dynamic FCs of DMN subsystems could be clustered into two recurring states. For dynamic FCs within MTL subsystem, WMH-MCI subjects exhibited longer mean dwell time (MDT) and higher reoccurrence fraction (RF) in a sparsely connected state (State 2). Altered MDT and RF in State 2 were negatively associated with IPS. Taken together, these findings indicated static/dynamic FC of DMN subsystems can provide relevant information on cognitive decline from different aspects, which provides a comprehensive view of subsystem mechanisms of DMN underlying WMH-related CI.

Temporal Stability of the Dynamic Resting-State Functional Brain Network: Current Measures, Clinical Research Progress, and Future Perspectives

Based on functional magnetic resonance imaging and multilayer dynamic network model, the brain network’s quantified temporal stability has shown potential in predicting altered brain functions. This manuscript aims to summarize current knowledge, clinical research progress, and future perspectives on brain network’s temporal stability. There are a variety of widely used measures of temporal stability such as the variance/standard deviation of dynamic functional connectivity strengths, the temporal variability, the flexibility (switching rate), and the temporal clustering coefficient, while there is no consensus to date which measure is the best. The temporal stability of brain networks may be associated with several factors such as sex, age, cognitive functions, head motion, circadian rhythm, and data preprocessing/analyzing strategies, which should be considered in clinical studies. Multiple common psychiatric disorders such as schizophrenia, major depressive disorder, and bipolar disorder have been found to be related to altered temporal stability, especially during the resting state; generally, both excessively decreased and increased temporal stabilities were thought to reflect disorder-related brain dysfunctions. However, the measures of temporal stability are still far from applications in clinical diagnoses for neuropsychiatric disorders partly because of the divergent results. Further studies with larger samples and in transdiagnostic (including schizoaffective disorder) subjects are warranted.

Inter-Network Brain Functional Connectivity in Adolescents Assigned Female at Birth Who Experience Gender Dysphoria

Gender dysphoria (GD) is characterized by distress due to an incongruence between experienced gender and sex assigned at birth. Brain functional connectivity in adolescents who experience GD may be associated with experienced gender (vs. assigned sex) and/or brain networks implicated in own-body perception. Furthermore, sexual orientation may be related to brain functional organization given commonalities in developmental mechanisms proposed to underpin GD and same-sex attractions. Here, we applied group independent component analysis to resting-state functional magnetic resonance imaging (rs-fMRI) BOLD timeseries data to estimate inter-network (i.e., between independent components) timeseries correlations, representing functional connectivity, in 17 GD adolescents assigned female at birth (AFAB) not receiving gender-affirming hormone therapy, 17 cisgender girls, and 15 cisgender boys (ages 12-17 years). Sexual orientation was represented by degree of androphilia-gynephilia and sexual attractions strength. Multivariate partial least squares analyses found that functional connectivity differed among cisgender boys, cisgender girls, and GD AFAB, with the largest difference between cisgender boys and GD AFAB. Regarding sexual orientation and age, the brain's intrinsic functional organization of GD AFAB was both similar to and different from cisgender girls, and both differed from cisgender boys. The pattern of group differences and the networks involved aligned with the hypothesis that brain functional organization is different among GD AFAB (vs. cisgender) adolescents, and certain aspects of this organization relate to brain areas implicated in own-body perception and self-referential thinking. Overall, brain functional organization of GD AFAB was generally more similar to that of cisgender girls than cisgender boys.

From YouTube to the brain: Transfer learning can improve brain-imaging predictions with deep learning

Deep learning has recently achieved best-in-class performance in several fields, including biomedical domains such as X-ray images. Yet, data scarcity poses a strict limit on training successful deep learning systems in many, if not most, biomedical applications, including those involving brain images. In this study, we translate state-of-the-art transfer learning techniques for single-subject prediction of simpler (sex and age) and more complex phenotypes (number of people in household, household income, fluid intelligence and smoking behavior). We fine-tuned 2D and 3D ResNet-18 convolutional neural networks for target phenotype predictions from brain images of ∼40,000 UK Biobank participants, after pretraining on YouTube videos from the Kinetics dataset and natural images from the ImageNet dataset. Transfer learning was effective on several phenotypes, especially sex and age classification. Additionally, transfer learning in particular outperformed deep learning models trained from scratch especially on smaller sample sizes. The out-of-sample performance using transfer learning from previously learned knowledge based on real-world images and videos could unlock the potential in many areas of imaging neuroscience where deep learning solutions are currently infeasible.

Sex-related differences in brain dynamism at rest as neural correlates of positive and negative valence system constructs

Clinical anxiety and depression are the most prevalent mental illnesses, likely representing maladaptive expressions of negative valence systems concerned with conditioned responses to fear, threat, loss, and frustrative nonreward. These conditions exhibit similar, striking sex/gender-related differences in onset, incidence, and severity for which the neural correlates are not yet established. In alarge sample of neurotypical young adults, we demonstrate that intrinsic brain dynamism metrics derived from sex-sensitive models of whole-brain network function are significantly associated with valence system traits. Surprisingly, we found that greater brain dynamism is strongly positively correlated to anxiety and depression traits in males, but almost wholly decoupled from traits for important cognitive control and reappraisal strategies associated with positive valence. Conversely, intrinsic brain dynamism is strongly positively coupled to drive, novelty-seeking and self-control in females with only rare or non-significant directional negative correlation with anxiety and depression traits. Our results suggest that the dynamic neural correlates of traits for valence, anxiety and depression are significantly different in males/men and females/women. These findings may relate to the known sex/gender-related differences in cognitive reappraisal of emotional experiences and clinical presentations of anxiety and depression, with potential relevance to gold standard therapies based on enhancing cognitive control strategies.

Sex differences in functional network dynamics observed using coactivation pattern analysis

Sex differences in the organization of large-scale resting-state brain networks have been identified using traditional static measures, which average functional connectivity over extended time periods. In contrast, emerging dynamic measures have the potential to define sex differences in network changes over time, providing additional understanding of neurobiological sex differences. To meet this goal, we used a Coactivation Pattern Analysis (CAP) using resting-state functional magnetic resonance imaging data from 181 males and 181 females from the Human Connectome Project. Significant main effects of sex were observed across two independent imaging sessions. Relative to males, females spent more total time in two transient network states (TNSs) spatially overlapping with the dorsal attention network and occipital/sensory-motor network. Greater time spent in these TNSs was related to females making more frequent transitions into these TNSs compared to males. In contrast, males spent more total time in TNSs spatially overlapping with the salience network, which was related to males staying for longer periods once entering these TNSs compared to females. State-to-state transitions also significantly differed between sexes: females transitioned more frequently from default mode network (DMN) states to the dorsal attention network state, whereas males transitioned more frequently from DMN states to salience network states. Results show that males and females spend differing amounts of time at rest in two distinct attention-related networks and show sex-specific transition patterns from DMN states into these attention-related networks. This work lays the groundwork for future investigations into the cognitive and behavioral implications of these sex-specific network dynamics.

Predicting Biological Gender and Intelligence From fMRI via Dynamic Functional Connectivity

OBJECTIVE

This paper explores the predictive capability of dynamic functional connectivity extracted from functional magnetic resonance imaging (fMRI) of the human brain, in contrast to static connectivity used in past research.

METHODS

Several state-of-the-art features extracted from static functional connectivity of the brain are employed to predict biological gender and intelligence using publicly available Human Connectome Project (HCP) database. Next, a novel tensor parallel factor (PARAFAC) decomposition model is proposed to decompose sequence of dynamic connectivity matrices into common connectivity components that are orthonormal to each other, common time-courses, and corresponding distinct subject-wise weights. The subject-wise loading of the components are employed to predict biological gender and intelligence using a random forest classifier (respectively, regressor) using 5-fold cross-validation.

RESULTS

The results demonstrate that dynamic functional connectivity can indeed classify biological gender with a high accuracy (0.94, where male identification accuracy was 0.87 and female identification accuracy was 0.97). It can also predict intelligence with less normalized mean square error (0.139 for fluid intelligence and 0.031 for fluid ability metrics) compared with other functional connectivity measures (the nearest mean square error were 0.147 and 0.037 for fluid intelligence and fluid ability metrics, respectively, using static connectivity approaches).

CONCLUSION

Our work is an important milestone for the understanding of non-stationary behavior of hemodynamic blood-oxygen level dependent (BOLD) signal in brain and how they are associated with biological gender and intelligence.

SIGNIFICANCE

The paper demonstrates that dynamic behavior of brain can contribute substantially towards forming a fingerprint of biological gender and intelligence.