Modelling neuroanatomical variation during childhood and adolescence with neighbourhood-preserving embedding

Predicting 'Brainage' in late childhood to adolescence (6-17yrs) using structural MRI, morphometric similarity, and machine learning

Brain development is regularly studied using structural MRI. Recently, studies have used a combination of statistical learning and large-scale imaging databases of healthy children to predict an individual's age from structural MRI. This data-driven, predicted 'Brainage' typically differs from the subjects chronological age, with this difference a potential measure of individual difference. Few studies have leveraged higher-order or connectomic representations of structural MRI data for this Brainage approach. We leveraged morphometric similarity as a network-level approach to structural MRI to generate predictive models of age. We benchmarked these novel Brainage approaches using morphometric similarity against more typical, single feature (i.e., cortical thickness) approaches. We showed that these novel methods did not outperform cortical thickness or cortical volume measures. All models were significantly biased by age, but robust to motion confounds. The main results show that, whilst morphometric similarity mapping may be a novel way to leverage additional information from a T1-weighted structural MRI beyond individual features, in the context of a Brainage framework, morphometric similarity does not provide more accurate predictions of age. Morphometric similarity as a network-level approach to structural MRI may be poorly positioned to study individual differences in brain development in healthy participants in this way.

Predicting 'Brainage' in the Developmental Period using Structural MRI, Morphometric Similarity, and Machine Learning

Brain development is regularly studied using structural MRI. Recently, studies have used a combination of statistical learning and large-scale imaging databases of healthy-children to predict an individual's age from structural MRI. This data-driven, 'brainage' typically differs from the subjects chronological age, with this difference a potential measure of individual difference. Few studies have leveraged higher-order or connectomic representations of structural MRI data for this brainage approach. We leveraged morphometric similarity as a network-level approach to structural MRI to generate predictive models of age. We benchmarked these novel brain-age approaches using morphometric similarity against more typical, single feature (i.e. cortical thickness) approaches. We showed that these novel methods did not outperform cortical thickness or cortical volume measures. All models were significantly biased by age, but robust to motion confounds. The main results show that, whilst morphometric similarity mapping may be a novel way to leverage additional information from a T1-weighted structural MRI beyond individual features, in the context of a brain-age framework, morphometric similarity does not explain more variance than individual structural features. Morphometric similarity as a network-level approach to structural MRI may be poorly positioned to study individual differences in brain development in healthy individuals.

Investigating brain structural maturation in children and adolescents born very preterm using the brain age framework

Very preterm (VP) birth is associated with an increased risk for later neurodevelopmental and behavioural challenges. Although the neurobiological underpinnings of such challenges continue to be explored, previous studies have reported brain volume and morphology alterations in children and adolescents born VP compared with full-term (FT)-born controls. How these alterations relate to the trajectory of brain maturation, with potential implications for later brain ageing, remains unclear. In this longitudinal study, we investigate the relationship between VP birth and brain development during childhood and adolescence. We construct a normative 'brain age' model to predict age over childhood and adolescence based on measures of brain cortical and subcortical volumes and cortical morphology from structural MRI of a dataset of typically developing children aged 3-21 years (n = 768). Using this model, we examined deviations from normative brain development in a separate dataset of children and adolescents born VP (<30 weeks' gestation) at two timepoints (ages 7 and 13 years) compared with FT-born controls (120 VP and 29 FT children at age 7 years; 140 VP and 47 FT children at age 13 years). Brain age delta (brain-predicted age minus chronological age) was, on average, higher in the VP group at both timepoints compared with controls, however this difference had a small to medium effect size and was not statistically significant. Variance in brain age delta was higher in the VP group compared with controls; this difference was significant at the 13-year timepoint. Within the VP group, there was little evidence of associations between brain age delta and perinatal risk factors or cognitive and motor outcomes. Under the brain age framework, our results may suggest that children and adolescents born VP have similar brain structural developmental trajectories to term-born peers between 7 and 13 years of age.

An efficient functional magnetic resonance imaging data reduction strategy using neighborhood preserving embedding algorithm

High dimensionality data have become common in neuroimaging fields, especially group‐level functional magnetic resonance imaging (fMRI) datasets. fMRI connectivity analysis is a widely used, powerful technique for studying functional brain networks to probe underlying mechanisms of brain function and neuropsychological disorders. However, data‐driven technique like independent components analysis (ICA), can yield unstable and inconsistent results, confounding the true effects of interest and hindering the understanding of brain functionality and connectivity. A key contributing factor to this instability is the information loss that occurs during fMRI data reduction. Data reduction of high dimensionality fMRI data in the temporal domain to identify the important information within group datasets is necessary for such analyses and is crucial to ensure the accuracy and stability of the outputs. In this study, we describe an fMRI data reduction strategy based on an adapted neighborhood preserving embedding (NPE) algorithm. Both simulated and real data results indicate that, compared with the widely used data reduction method, principal component analysis, the NPE‐based data reduction method (a) shows superior performance on efficient data reduction, while enhancing group‐level information, (b) develops a unique stratagem for selecting components based on an adjacency graph of eigenvectors, (c) generates more reliable and reproducible brain networks under different model orders when the outputs of NPE are used for ICA, (d) is more sensitive to revealing task‐evoked activation for task fMRI, and (e) is extremely attractive and powerful for the increasingly popular fast fMRI and very large datasets.

Viewing Pictures Triggers Rapid Morphological Enlargement in the Human Visual Cortex

Measuring brain morphology with non-invasive structural magnetic resonance imaging is common practice, and can be used to investigate neuroplasticity. Brain morphology changes have been reported over the course of weeks, days, and hours in both animals and humans. If such short-term changes occur even faster, rapid morphological changes while being scanned could have important implications. In a randomized within-subject study on 47 healthy individuals, two high-resolution T1-weighted anatomical images were acquired (á 263 s) per individual. The images were acquired during passive viewing of pictures or a fixation cross. Two common pipelines for analyzing brain images were used: voxel-based morphometry on gray matter (GM) volume and surface-based cortical thickness. We found that the measures of both GM volume and cortical thickness showed increases in the visual cortex while viewing pictures relative to a fixation cross. The increase was distributed across the two hemispheres and significant at a corrected level. Thus, brain morphology enlargements were detected in less than 263 s. Neuroplasticity is a far more dynamic process than previously shown, suggesting that individuals’ current mental state affects indices of brain morphology. This needs to be taken into account in future morphology studies and in everyday clinical practice.

Individual variation underlying brain age estimates in typical development

Typical brain development follows a protracted trajectory throughout childhood and adolescence. Deviations from typical growth trajectories have been implicated in neurodevelopmental and psychiatric disorders. Recently, the use of machine learning algorithms to model age as a function of structural or functional brain properties has been used to examine advanced or delayed brain maturation in healthy and clinical populations. Termed 'brain age', this approach often relies on complex, nonlinear models that can be difficult to interpret. In this study, we use model explanation methods to examine the cortical features that contribute to brain age modelling on an individual basis. In a large cohort of n = 768 typically-developing children (aged 3-21 years), we build models of brain development using three different machine learning approaches. We employ SHAP, a model-agnostic technique to identify sample-specific feature importance, to identify regional cortical metrics that explain errors in brain age prediction. We find that, on average, brain age prediction and the cortical features that explain model predictions are consistent across model types and reflect previously reported patterns of regions brain development. However, while several regions are found to contribute to brain age prediction error, we find little spatial correspondence between individual estimates of feature importance, even when matched for age, sex and brain age prediction error. We also find no association between brain age error and cognitive performance in this typically-developing sample. Overall, this study shows that, while brain age estimates based on cortical development are relatively robust and consistent across model types and preprocessing strategies, significant between-subject variation exists in the features that explain erroneous brain age predictions on an individual level.

Accelerated brain aging predicts impulsivity and symptom severity in depression

Multiple structural and functional neuroimaging measures vary over the course of the lifespan and can be used to predict chronological age. Accelerated brain aging, as quantified by deviations in the MRI-based predicted age with respect to chronological age, is associated with risk for neurodegenerative conditions, bipolar disorder, and mortality. Whether age-related changes in resting-state functional connectivity are accelerated in major depressive disorder (MDD) is unknown, and, if so, it is unclear if these changes contribute to specific cognitive weaknesses that often occur in MDD. Here, we delineated age-related functional connectivity changes in a large sample of normal control subjects and tested whether brain aging is accelerated in MDD. Furthermore, we tested whether accelerated brain aging predicts individual differences in cognitive function. We trained a support vector regression model predicting age using resting-state functional connectivity in 710 healthy adults aged 18-89. We applied this model trained on normal aging subjects to a sample of actively depressed MDD participants (n = 109). The difference between predicted brain age and chronological age was 2.11 years greater (p = 0.015) in MDD patients compared to control participants. An older MDD brain age was significantly associated with increased impulsivity and, in males, increased depressive severity. Unexpectedly, accelerated brain aging was also associated with increased placebo response in a sham-controlled trial of high-frequency repetitive transcranial magnetic stimulation targeting the dorsomedial prefrontal cortex. Our results indicate that MDD is associated with accelerated brain aging, and that accelerated aging is selectively associated with greater impulsivity and depression severity.

Predicting brain age with complex networks: From adolescence to adulthood

In recent years, several studies have demonstrated that machine learning and deep learning systems can be very useful to accurately predict brain age. In this work, we propose a novel approach based on complex networks using 1016 T1-weighted MRI brain scans (in the age range 7-64years). We introduce a structural connectivity model of the human brain: MRI scans are divided in rectangular boxes and Pearson's correlation is measured among them in order to obtain a complex network model. Brain connectivity is then characterized through few and easy-to-interpret centrality measures; finally, brain age is predicted by feeding a compact deep neural network. The proposed approach is accurate, robust and computationally efficient, despite the large and heterogeneous dataset used. Age prediction accuracy, in terms of correlation between predicted and actual age r=0.89and Mean Absolute Error MAE =2.19years, compares favorably with results from state-of-the-art approaches. On an independent test set including 262 subjects, whose scans were acquired with different scanners and protocols we found MAE =2.52. The only imaging analysis steps required in the proposed framework are brain extraction and linear registration, hence robust results are obtained with a low computational cost. In addition, the network model provides a novel insight on aging patterns within the brain and specific information about anatomical districts displaying relevant changes with aging.

Brain status modeling with non-negative projective dictionary learning

Accurate prediction of individuals' brain age is critical to establish a baseline for normal brain development. This study proposes to model brain development with a novel non-negative projective dictionary learning (NPDL) approach, which learns a discriminative representation of multi-modal neuroimaging data for predicting brain age. Our approach encodes the variability of subjects in different age groups using separate dictionaries, projecting features into a low-dimensional manifold such that information is preserved only for the corresponding age group. The proposed framework improves upon previous discriminative dictionary learning methods by incorporating orthogonality and non-negativity constraints, which remove representation redundancy and perform implicit feature selection. We study brain development on multi-modal brain imaging data from the PING dataset (N = 841, age = 3-21 years). The proposed analysis uses our NDPL framework to predict the age of subjects based on cortical measures from T1-weighted MRI and connectome from diffusion weighted imaging (DWI). We also investigate the association between age prediction and cognition, and study the influence of gender on prediction accuracy. Experimental results demonstrate the usefulness of NDPL for modeling brain development.

Multimodal Sparse Classifier for Adolescent Brain Age Prediction

The study of healthy brain development helps to better understand both brain transformation and connectivity patterns, which happen during childhood to adulthood. This study presents a sparse machine learning solution across whole-brain functional connectivity measures of three datasets, derived from resting state functional magnetic resonance imaging (rs-fMRI) and two task fMRI data including a working memory n-back task (nb-fMRI) and an emotion identification task (em-fMRI). The fMRI data are collected from the Philadelphia Neurodevelopmental Cohort (PNC) for the prediction of brain age in adolescents. Due to extremely large variable-to-instance ratio of PNC data, a high-dimensional matrix with several irrelevant and highly correlated features is generated, and hence a sparse learning approach is necessary to extract effective features from fMRI data. We propose a sparse learner based on the residual errors along the estimation of an inverse problem for extreme learning machine (ELM). Our proposed method is able to overcome the overlearning problem by pruning several redundant features and their corresponding output weights. The proposed multimodal sparse ELM classifier based on residual errors is highly competitive in terms of classification accuracy compared to its counterparts such as conventional ELM, and sparse Bayesian learning ELM.

Charting shared developmental trajectories of cortical thickness and structural connectivity in childhood and adolescence

The cortex is organised into broadly hierarchical functional systems with distinct neuroanatomical characteristics reflected by macroscopic measures of cortical morphology. Diffusion-weighted magnetic resonance imaging allows the delineation of areal connectivity, changes to which reflect the ongoing maturation of white matter tracts. These developmental processes are intrinsically linked with timing coincident with the development of cognitive function. In this study, we use a data-driven multivariate approach, nonnegative matrix factorisation, to define cortical regions that co-vary together across a large paediatric cohort (n = 456) and are associated with specific subnetworks of cortical connectivity. We find that age between 3 and 21 years is associated with accelerated cortical thinning in frontoparietal regions, whereas relative thinning of primary motor and sensory regions is slower. Together, the subject-specific weights of the derived set of cortical components can be combined to predict chronological age. Structural connectivity networks reveal a relative increase in strength in connection within, as opposed to between hemispheres that vary in line with cortical changes. We confirm our findings in an independent sample.

Accelerating research on biological aging and mental health: Current challenges and future directions

Aging is associated with complex biological changes that can be accelerated, slowed, or even temporarily reversed by biological and non-biological factors. This article focuses on the link between biological aging, psychological stressors, and mental illness. Rather than comprehensively reviewing this rapidly expanding field, we highlight challenges in this area of research and propose potential strategies to accelerate progress in this field. This effort requires the interaction of scientists across disciplines - including biology, psychiatry, psychology, and epidemiology; and across levels of analysis that emphasize different outcome measures - functional capacity, physiological, cellular, and molecular. Dialogues across disciplines and levels of analysis naturally lead to new opportunities for discovery but also to stimulating challenges. Some important challenges consist of 1) establishing the best objective and predictive biological age indicators or combinations of indicators, 2) identifying the basis for inter-individual differences in the rate of biological aging, and 3) examining to what extent interventions can delay, halt or temporarily reverse aging trajectories. Discovering how psychological states influence biological aging, and vice versa, has the potential to create novel and exciting opportunities for healthcare and possibly yield insights into the fundamental mechanisms that drive human aging.

Sex Effects on Development of Brain Structure and Executive Functions: Greater Variance than Mean Effects

Although male brains have consistently reported to be 8-10% larger than female brains, it remains not well understood whether there are differences between sexes (average or variance) in developmental trajectories. Furthermore, if sex differences in average brain growth or variance are observed, it is unknown whether these sex differences have behavioral relevance. The present longitudinal study aimed to unravel sex effects in cortical brain structure, development, and variance, in relation to the development of educationally relevant cognitive domains and executive functions (EFs). This was assessed with three experimental tasks including working memory, reading comprehension, and fluency. In addition, real-life aspects of EF were assessed with self- and parent-reported Behavior Rating Inventory of Executive Function scores. The full data set included 271 participants (54% female) aged between 8 and 29 years of which three waves were collected at 2-year intervals, resulting in 680 T1-weighted MRI scans and behavioral measures. Analyses of average trajectories confirmed general age-related patterns of brain development but did not support the hypothesis of sex differences in brain development trajectories, except for left banks STS where boys had a steeper decline in surface area than girls. Also, our brain age prediction model (including 270 brain measures) did not indicate delayed maturation in boys compared with girls. Interestingly, support was found for greater variance in male brains than female brains in both structure and development, consistent with prior cross-sectional studies. Behaviorally, boys performed on average better on a working memory task with a spatial aspect and girls performed better on a reading comprehension task, but there was no relation between brain development and cognitive performance, neither for average brain measures, brain age, or variance measures. Taken together, we confirmed the hypothesis of greater males within-group variance in brain structures compared with females, but these were not related to EF. The sex differences observed in EF were not related to brain development, possibly suggesting that these are related to experiences and strategies rather than biological development.

Unbiased age-specific structural brain atlases for Chinese pediatric population

In magnetic resonance (MR) imaging studies of child brain development, structural brain atlases usually serve as important references for the pediatric population, in which individual images are spatially normalized into a common or standard stereotactic space. However, the popular existing pediatric brain atlases (e.g., National Institutes of Health pediatric atlases, NIH-PD) are mostly based on MR images obtained from Caucasian populations and thus are not ideal for the characterization of the brains of Chinese children due to neuroanatomical differences related to genetic and environmental factors. Here, we use an unbiased template construction algorithm to create a set of age-specific Chinese pediatric (CHN-PD) atlases based on high-quality T1-and T2-weighted MR images from 328 cognitively normal Chinese children aged 6-12 years. The CHN-PD brain atlases include asymmetric and symmetric templates, sex-specific templates and tissue probability templates, and contain multiple age-specific templates at one-year intervals. A direct comparison of the CHN-PD and NIH-PD atlases reveals dramatic anatomical differences mainly in the bilateral frontal and parietal regions. After applying the CHN-PD and NIH-PD atlases to two independent Chinese pediatric datasets (N = 114 and N = 71), we find that the CHN-PD atlases result in significantly higher accuracy than the NIH-PD atlases in both predicting "brain age" and guiding brain tissue segmentation. These results suggest that the CHN-PD brain atlases are necessary for studies of the typical and atypical development of the Chinese pediatric population. These CHN-PD atlases have been released on the Neuroimaging Informatics Tools and Resources Clearinghouse (NITRC) website (https://www.nitrc.org/projects/chn-pd).