scholarly journals The Vital Role of Central Executive Network in Brain Age: Evidence From Machine Learning and Transcriptional Signatures

2021 ◽  
Vol 15 ◽  
Author(s):  
Keke Fang ◽  
Shaoqiang Han ◽  
Yuming Li ◽  
Jing Ding ◽  
Jilian Wu ◽  
...  
Keyword(s):  
Machine Learning ◽  
Gaussian Process Regression ◽  
Enrichment Analysis ◽  
Brain Network ◽  
Vital Role ◽  
Central Executive ◽  
Age Related ◽  
Central Executive Network ◽  
Brain Age ◽  
Age Prediction

Recent studies combining neuroimaging with machine learning methods successfully infer an individual’s brain age, and its discrepancy with the chronological age is used to identify age-related diseases. However, which brain networks play decisive roles in brain age prediction and the underlying biological basis of brain age remain unknown. To answer these questions, we estimated an individual’s brain age in the Southwest University Adult Lifespan Dataset (N = 492) from the gray matter volumes (GMV) derived from T1-weighted MRI scans by means of Gaussian process regression. Computational lesion analysis was performed to determine the importance of each brain network in brain age prediction. Then, we identified brain age-related genes by using prior brain-wide gene expression data, followed by gene enrichment analysis using Metascape. As a result, the prediction model successfully inferred an individual’s brain age and the computational lesion prediction results identified the central executive network as a vital network in brain age prediction (Steiger’s Z = 2.114, p = 0.035). In addition, the brain age-related genes were enriched in Gene Ontology (GO) processes/Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways grouped into numbers of clusters, such as regulation of iron transmembrane transport, synaptic signaling, synapse organization, retrograde endocannabinoid signaling (e.g., dopaminergic synapse), behavior (e.g., memory and associative learning), neurotransmitter secretion, and dendrite development. In all, these results reveal that the GMV of the central executive network played a vital role in predicting brain age and bridged the gap between transcriptome and neuroimaging promoting an integrative understanding of the pathophysiology of brain age.

scholarly journals Brain age prediction in schizophrenia: Does the choice of machine learning algorithm matter?

2021 ◽  
Vol 310 ◽  
pp. 111270
Author(s):  
Won Hee Lee ◽  
Mathilde Antoniades ◽  
Hugo G Schnack ◽  
Rene S. Kahn ◽  
Sophia Frangou
Keyword(s):  
Machine Learning ◽  
Learning Algorithm ◽  
Machine Learning Algorithm ◽  
Brain Age ◽  
Age Prediction

scholarly journals Brain age prediction: A comparison between machine learning models using region‐ and voxel‐based morphometric data

2021 ◽  
Author(s):  
Lea Baecker ◽  
Jessica Dafflon ◽  
Pedro F. Costa ◽  
Rafael Garcia‐Dias ◽  
Sandra Vieira ◽  
...  
Keyword(s):  
Machine Learning ◽  
Learning Models ◽  
Morphometric Data ◽  
Brain Age ◽  
Machine Learning Models ◽  
Age Prediction

scholarly journals Individual variation underlying brain age estimates in typical development

2020 ◽  
Author(s):  
Gareth Ball ◽  
Claire E Kelly ◽  
Richard Beare ◽  
Marc L Seal
Keyword(s):  
Machine Learning ◽  
Brain Development ◽  
Machine Learning Algorithms ◽  
Brain Maturation ◽  
Childhood And Adolescence ◽  
Typically Developing ◽  
Feature Importance ◽  
Brain Age ◽  
Age Estimates ◽  
Age Prediction

AbstractTypical 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 estimate 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 regional brain development. However, while several regions are found to contribute to brain age prediction, 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.

scholarly journals Systematic Evaluation of Machine Learning Algorithms for Neuroanatomically-Based Age Prediction in Youth

2021 ◽  
Author(s):  
Mandana Modabbernia ◽  
Heather C Whalley ◽  
David Glahn ◽  
Paul M. Thompson ◽  
Rene S. Kahn ◽  
...  
Keyword(s):  
Machine Learning ◽  
Computational Efficiency ◽  
Learning Algorithms ◽  
Machine Learning Algorithms ◽  
Sensitivity Analyses ◽  
Gradient Boosting ◽  
Support Vector ◽  
Age Related ◽  
Extreme Gradient Boosting ◽  
Brain Age

Application of machine learning algorithms to structural magnetic resonance imaging (sMRI) data has yielded behaviorally meaningful estimates of the biological age of the brain (brain-age). The choice of the machine learning approach in estimating brain-age in children and adolescents is important because age-related brain changes in these age-groups are dynamic. However, the comparative performance of the multiple machine learning algorithms available has not been systematically appraised. To address this gap, the present study evaluated the accuracy (Mean Absolute Error; MAE) and computational efficiency of 21 machine learning algorithms using sMRI data from 2,105 typically developing individuals aged 5 to 22 years from five cohorts. The trained models were then tested in an independent holdout datasets, comprising 4,078 pre-adolescents (aged 9-10 years). The algorithms encompassed parametric and nonparametric, Bayesian, linear and nonlinear, tree-based, and kernel-based models. Sensitivity analyses were performed for parcellation scheme, number of neuroimaging input features, number of cross-validation folds, and sample size. The best performing algorithms were Extreme Gradient Boosting (MAE of 1.25 years for females and 1.57 years for males), Random Forest Regression (MAE of 1.23 years for females and 1.65 years for males) and Support Vector Regression with Radial Basis Function Kernel (MAE of 1.47 years for females and 1.72 years for males) which had acceptable and comparable computational efficiency. Findings of the present study could be used as a guide for optimizing methodology when quantifying age-related changes during development.

scholarly journals Brain age prediction using fMRI network coupling in youths and associations with psychiatric symptoms

2021 ◽  
Author(s):  
Martina J. Lund ◽  
Dag Alnæs ◽  
Ann-Marie de Lange ◽  
Ole A. Andreassen ◽  
Lars T. Westlye ◽  
...  
Keyword(s):  
Functional Connectivity ◽  
Psychiatric Symptoms ◽  
Model Performance ◽  
Symptom Burden ◽  
Brain Network ◽  
Brain Maturation ◽  
Age Gap ◽  
Age Range ◽  
Brain Age ◽  
Age Prediction

AbstractObjectiveMagnetic resonance imaging (MRI) has shown that estimated brain age is deviant from chronological age in various common brain disorders. Brain age estimation could be useful for investigating patterns of brain maturation and integrity, aiding to elucidate brain mechanisms underlying these heterogeneous conditions. Here, we examined functional brain age in two large samples of children and adolescents and its relation to mental health.MethodsWe used resting-state fMRI data from the Philadelphia Neurodevelopmental Cohort (PNC; n=1126, age range 8-22 years) to estimate functional connectivity between brain networks, and utilized these as features for brain age prediction. We applied the prediction model to 1387 individuals (age range 8-22 years) in the Healthy Brain Network sample (HBN). In addition, we estimated brain age in PNC using a cross-validation framework. Next, we tested for associations between brain age gap and various aspects of psychopathology and cognitive performance.ResultsOur model was able to predict age in the independent test samples, with a model performance of r=0.54 for the HBN test set, supporting consistency in functional connectivity patterns between samples and scanners. Linear models revealed a significant association between brain age gap and psychopathology in PNC, where individuals with a lower estimated brain age, had a higher overall symptom burden. These associations were not replicated in HBN.DiscussionOur findings support the use of brain age prediction from fMRI-based connectivity. While requiring further extensions and validations, the approach may be instrumental for detecting brain phenotypes related to intrinsic connectivity and could assist in characterizing risk in non-typically developing populations.

scholarly journals Brain age prediction in schizophrenia: does the choice of machine learning algorithm matter?

2020 ◽  
Author(s):  
Won Hee Lee ◽  
Mathilde Antoniades ◽  
Hugo G Schnack ◽  
Rene S. Kahn ◽  
Sophia Frangou
Keyword(s):  
Machine Learning ◽  
Machine Learning Algorithms ◽  
Chronological Age ◽  
Lasso Regression ◽  
Healthy Individuals ◽  
Ols Regression ◽  
Regional Regression ◽  
The Mean ◽  
Brain Age ◽  
Age Prediction

AbstractBackgroundSchizophrenia has been associated with lifelong deviations in the normative trajectories of brain structure. These deviations can be captured using the brain-predicted age difference (brainPAD), which is the difference between the biological age of an individual’s brain, as inferred from neuroimaging data, and their chronological age. Various machine learning algorithms are currently used for this purpose but their comparative performance has yet to be systematically evaluated.MethodsSix linear regression algorithms, ordinary least squares (OLS) regression, ridge regression, least absolute shrinkage and selection operator (Lasso) regression, elastic-net regression, linear support vector regression (SVR), and relevance vector regression (RVR), were applied to brain structural data acquired on the same 3T scanner using identical sequences from patients with schizophrenia (n=90) and healthy individuals (n=200). The performance of each algorithm was quantified by the mean absolute error (MAE) and the correlation (R) between predicted brain-age and chronological age. The inter-algorithm similarity in predicted brain-age, brain regional regression weights and brainPAD were compared using correlation analyses and hierarchical clustering.ResultsIn patients with schizophrenia, ridge regression, Lasso regression, elastic-net regression, and RVR performed very similarly and showed a high degree of correlation in predicted brain-age (R>0.94) and brain regional regression weights (R>0.66). By contrast, OLS regression, which was the only algorithm without a penalty term, performed markedly worse and showed a lower similarity with the other algorithms. The mean brainPAD was higher in patients than in healthy individuals but varied by algorithm from 3.8 to 5.2 years although all analyses were performed on the same dataset.ConclusionsLinear machine learning algorithms, with the exception of OLS regression, have comparable performance for age prediction on the basis of a combination of cortical and subcortical structural measures. However, algorithm choice introduced variation in brainPAD estimation, and therefore represents an important source of inter-study variability.

scholarly journals Machine learning for brain age prediction: Introduction to methods and clinical applications

EBioMedicine ◽  
2021 ◽  
Vol 72 ◽  
pp. 103600
Author(s):  
Lea Baecker ◽  
Rafael Garcia-Dias ◽  
Sandra Vieira ◽  
Cristina Scarpazza ◽  
Andrea Mechelli
Keyword(s):  
Machine Learning ◽  
Clinical Applications ◽  
Brain Age ◽  
Age Prediction

scholarly journals Association versus Prediction: the impact of cortical surface smoothing and parcellation on brain age

2020 ◽  
Author(s):  
Yashar Zeighami ◽  
Alan C. Evans
Keyword(s):  
Cortical Thickness ◽  
Prediction Accuracy ◽  
Negative Relationship ◽  
Correct Choice ◽  
Age Related ◽  
Smoothing Effects ◽  
The Impact ◽  
The Relationship ◽  
Brain Age ◽  
Age Prediction

AbstractAssociation and prediction studies of the brain target the biological consequences of aging and their impact on brain function. Such studies are conducted using different smoothing levels and parcellations at the preprocessing stage, on which their results are dependent. However, the impact of these parameters on the relationship between association values and prediction accuracy is not established. In this study, we used cortical thickness and its relationship with age to investigate how different smoothing and parcellation levels affect the detection of age-related brain correlates as well as brain age prediction accuracy. Our main measures were resel numbers - resolution elements - and age-related variance explained. Using these common measures enabled us to directly compare parcellation and smoothing effects in both association and prediction studies. In our sample of N=608 participants with age range 18-88, we evaluated age-related cortical thickness changes as well as brain age prediction. We found a negative relationship between prediction performance and correlation values for both parameters. Our results also quantify the relationship between delta age estimates obtained based on different processing parameters. Furthermore, with the direct comparison of the two approaches, we highlight the importance of correct choice of smoothing and parcellation parameters in each task, and how they can affect the results of the analysis in opposite directions.

scholarly journals Fast qualitY conTrol meThod foR derIved diffUsion Metrics (YTTRIUM) in big data analysis: UK Biobank 18608 example

2020 ◽  
Author(s):  
Ivan I. Maximov ◽  
Dennis van der Meer ◽  
Ann-Marie de Lange ◽  
Tobias Kaufmann ◽  
Alexey Shadrin ◽  
...  
Keyword(s):  
Machine Learning ◽  
Quality Control ◽  
Large Scale ◽  
Control Method ◽  
Large Population ◽  
Uk Biobank ◽  
Quality Control Method ◽  
Brain Age ◽  
Age Prediction

AbstractDeriving reliable information about the structural and functional architecture of the brain in vivo is critical for the clinical and basic neurosciences. In the new era of large population-based datasets, when multiple brain imaging modalities and contrasts are combined in order to reveal latent brain structural patterns and associations with genetic, demographic and clinical information, automated and stringent quality control (QC) procedures are important. Diffusion magnetic resonance imaging (dMRI) is a fertile imaging technique for probing and visualising brain tissue microstructure in vivo, and has been included in most standard imaging protocols in large-scale studies. Due to its sensitivity to subject motion and technical artefacts, automated QC procedures prior to statistical analyses of dMRI data are required to minimise the influence of noise and artefacts. Here, we introduce Fast qualitY conTrol meThod foR derIved diffUsion Metrics (YTTRIUM), a computationally efficient QC method utilising structural similarity to evaluate image quality and mean diffusion metrics. As an example, we applied YTTRIUM in the context of tract-based spatial statistics to assess associations between age and kurtosis imaging and white matter tract integrity in UK Biobank data (n = 18,608). In order to assess the influence of outliers on results obtained using machine learning approaches, we tested the effects of applying YTTRIUM on brain age prediction. We demonstrated that the proposed QC pipeline represents an efficient approach for identifying poor quality datasets and artifacts and increase the accuracy of machine learning based brain age prediction.

Sign in / Sign up

Export Citation Format

Share Document