scholarly journals Predicting age from cortical structure across the lifespan

2018 ◽  
Author(s):  
Christopher R. Madan ◽  
Elizabeth A. Kensinger
Keyword(s):  
Fractal Dimensionality ◽  
Brain Morphology ◽  
Training Dataset ◽  
Anatomical Landmarks ◽  
Cortical Structure ◽  
Age Related ◽  
Median Error ◽  
Structural Measures ◽  
Cortical Morphology ◽  
Age Prediction

AbstractDespite inter-individual differences in cortical structure, cross-sectional and longitudinal studies have demonstrated a large degree of population-level consistency in age-related differences in brain morphology. The present study assessed how accurately an individual’s age could be predicted by estimates of cortical morphology, comparing a variety of structural measures, including thickness, gyrification, and fractal dimensionality. Structural measures were calculated across up to seven different parcellation approaches, ranging from 1 region to 1000 regions. The age-prediction framework was trained using morphological measures obtained from T1-weighted MRI volumes collected from multiple sites, yielding a training dataset of 1056 healthy adults, aged 18-97. Age predictions were calculated using a machine-learning approach that incorporated non-linear differences over the lifespan. In two independent, held-out test samples, age predictions had a median error of 6-7 years. Age predictions were best when using a combination of cortical metrics, both thickness and fractal dimensionality. Overall, the results reveal that age-related differences in brain structure are systematic enough to enable reliable age prediction based on metrics of cortical morphology.Graphical AbstractSeveral measures of cortical structure differ in relation to age. We examined the cortical granularity of these differences across seven parcellation approaches, from a 1 region (unparcellated cortical ribbon) to 1000 regions (patches with boundaries informed by anatomical landmarks), and three measures: thickness, gyrification, and fractal dimensionality. Rather than merely examining age-related relationships, we examined how these parcellations and measures can be used topredictage.

scholarly journals Human Age Prediction Based on DNA Methylation Using a Gradient Boosting Regressor

Genes ◽  
2018 ◽  
Vol 9 (9) ◽  
pp. 424 ◽  
Author(s):  
Xingyan Li ◽  
Weidong Li ◽  
Yan Xu
Keyword(s):  
Dna Methylation ◽  
Pearson Correlation ◽  
Training Data ◽  
Training Dataset ◽  
Gradient Boosting ◽  
Independent Dataset ◽  
Blood Samples ◽  
Cpg Sites ◽  
Age Related ◽  
Age Prediction

All tissues of organisms will become old as time goes on. In recent years, epigenetic investigations have found that there is a close correlation between DNA methylation and aging. With the development of DNA methylation research, a quantitative statistical relationship between DNA methylation and different ages was established based on the change rule of methylation with age, it is then possible to predict the age of individuals. All the data in this work were retrieved from the Illumina HumanMethylation BeadChip platform (27K or 450K). We analyzed 16 sets of healthy samples and 9 sets of diseased samples. The healthy samples included a total of 1899 publicly available blood samples (0–103 years old) and the diseased samples included 2395 blood samples. Six age-related CpG sites were selected through calculating Pearson correlation coefficients between age and DNA methylation values. We built a gradient boosting regressor model for these age-related CpG sites. 70% of the data was randomly selected as training data and the other 30% as independent data in each dataset for 25 runs in total. In the training dataset, the healthy samples showed that the correlation between predicted age and DNA methylation was 0.97, and the mean absolute deviation (MAD) was 2.72 years. In the independent dataset, the MAD was 4.06 years. The proposed model was further tested using the diseased samples. The MAD was 5.44 years for the training dataset and 7.08 years for the independent dataset. Furthermore, our model worked well when it was applied to saliva samples. These results illustrated that the age prediction based on six DNA methylation markers is very effective using the gradient boosting regressor.

scholarly journals Advances in studying brain morphology: The benefits of open-access data

2017 ◽  
Author(s):  
Christopher R Madan
Keyword(s):  
Open Access ◽  
Brain Structure ◽  
Large Scale ◽  
Open Data ◽  
Brain Morphology ◽  
Subcortical Structures ◽  
Open Access Data ◽  
Age Related ◽  
Neuroimaging Data ◽  
Access Data

Until recently, neuroimaging data for a research study needed to be collected within one’s own lab. However, when studying inter-individual differences in brain structure, a large sample of participants is necessary. Given the financial costs involved in collecting neuroimaging data from hundreds or thousands of participants, large-scale studies of brain morphology could previously only be conducted by well-funded laboratories with access to MRI facilities and to large samples of participants. With the advent of broad open-access data-sharing initiatives, this has recently changed–here the primary goal of the study is to collect large datasets to be shared, rather than sharing of the data as an afterthought. This paradigm shift is evident as increase in the pace of discovery, leading to a rapid rate of advances in our characterization of brain structure. The utility of open-access brain morphology data is numerous, ranging from observing novel patterns of age-related differences in subcortical structures to the development of more robust cortical parcellation atlases, with these advances being translatable to improved methods for characterizing clinical disorders (see Figure 1 for an illustration). Moreover, structural MRIs are generally more robust than functional MRIs, relative to potential artifacts and in being not task-dependent, resulting in large potential yields. While the benefits of open-access data have been discussed more broadly within the field of cognitive neuroscience elsewhere (Gilmore et al., 2017; Poldrack and Gorgolewski, 2014; Van Horn and Gazzaniga, 2013; Voytek, 2016), as well as in other fields (Ascoli et al., 2017; Choudhury et al., 2014; Davies et al., 2017), the current paper is focused specifically on the implications of open data to brain morphology research.

scholarly journals Brain charts for the human lifespan

2021 ◽  
Author(s):  
Richard A.I. Bethlehem ◽  
Jakob Seidlitz ◽  
Simon R. White ◽  
Jacob W. Vogel ◽  
Kevin M. Anderson ◽  
...  
Keyword(s):  
Individual Differences ◽  
Clinical Studies ◽  
Collaborative Study ◽  
Basic Research ◽  
Brain Morphology ◽  
Longitudinal Assessment ◽  
Human Lifespan ◽  
Genetic Heritability ◽  
Age Related ◽  
Mri Study

Over the past 25 years, neuroimaging has become a ubiquitous tool in basic research and clinical studies of the human brain. However, there are no reference standards against which to anchor measures of individual differences in brain morphology, in contrast to growth charts for traits such as height and weight. Here, we built an interactive online resource (www.brainchart.io) to quantify individual differences in brain structure from any current or future magnetic resonance imaging (MRI) study, against models of expected age-related trends. With the goal of basing these on the largest and most inclusive dataset, we aggregated MRI data spanning 115 days post-conception through 100 postnatal years, totaling 122,123 scans from 100,071 individuals in over 100 studies across 6 continents. When quantified as centile scores relative to the reference models, individual differences show high validity with non-MRI brain growth estimates and high stability across longitudinal assessment. Centile scores helped identify previously unreported brain developmental milestones and demonstrated increased genetic heritability compared to non-centiled MRI phenotypes. Crucially for the study of brain disorders, centile scores provide a standardised and interpretable measure of deviation that reveals new patterns of neuroanatomical differences across neurological and psychiatric disorders emerging during development and ageing. In sum, brain charts for the human lifespan are an essential first step towards robust, standardised quantification of individual variation and for characterizing deviation from age-related trends. Our global collaborative study provides such an anchorpoint for basic neuroimaging research and will facilitate implementation of research-based standards in clinical studies.

scholarly journals MapReduce-Based Parallel Genetic Algorithm for CpG-Site Selection in Age Prediction

Genes ◽  
2019 ◽  
Vol 10 (12) ◽  
pp. 969
Author(s):  
Zahra Momeni ◽  
Mohammad Saniee Abadeh
Keyword(s):  
Genetic Algorithm ◽  
Fitness Function ◽  
Gradient Boosting ◽  
Range Data ◽  
Parallel Genetic Algorithm ◽  
Data Parallel ◽  
Cpg Sites ◽  
Age Related ◽  
Age Range ◽  
Age Prediction

Genomic biomarkers such as DNA methylation (DNAm) are employed for age prediction. In recent years, several studies have suggested the association between changes in DNAm and its effect on human age. The high dimensional nature of this type of data significantly increases the execution time of modeling algorithms. To mitigate this problem, we propose a two-stage parallel algorithm for selection of age related CpG-sites. The algorithm first attempts to cluster the data into similar age ranges. In the next stage, a parallel genetic algorithm (PGA), based on the MapReduce paradigm (MR-based PGA), is used for selecting age-related features of each individual age range. In the proposed method, the execution of the algorithm for each age range (data parallel), the evaluation of chromosomes (task parallel) and the calculation of the fitness function (data parallel) are performed using a novel parallel framework. In this paper, we consider 16 different healthy DNAm datasets that are related to the human blood tissue and that contain the relevant age information. These datasets are combined into a single unioned set, which is in turn randomly divided into two sets of train and test data with a ratio of 7:3, respectively. We build a Gradient Boosting Regressor (GBR) model on the selected CpG-sites from the train set. To evaluate the model accuracy, we compared our results with state-of-the-art approaches that used these datasets, and observed that our method performs better on the unseen test dataset with a Mean Absolute Deviation (MAD) of 3.62 years, and a correlation (R2) of 95.96% between age and DNAm. In the train data, the MAD and R2 are 1.27 years and 99.27%, respectively. Finally, we evaluate our method in terms of the effect of parallelization in computation time. The algorithm without parallelization requires 4123 min to complete, whereas the parallelized execution on 3 computing machines having 32 processing cores each, only takes a total of 58 min. This shows that our proposed algorithm is both efficient and scalable.

scholarly journals The Effects of Lutein and Zeaxanthin Supplementation on Brain Morphology in Older Adults: A Randomized, Controlled Trial

2019 ◽  
Vol 2019 ◽  
pp. 1-11
Author(s):  
Catherine M. Mewborn ◽  
Cutter A. Lindbergh ◽  
B. Randy Hammond ◽  
Lisa M. Renzi-Hammond ◽  
L. Stephen Miller
Keyword(s):  
Older Adults ◽  
Placebo Group ◽  
White Matter ◽  
Controlled Trial ◽  
Community Dwelling ◽  
Brain Morphology ◽  
Age Related ◽  
Supplement Group ◽  
One Year ◽  
Structural Brain

A growing literature emphasizes the importance of lifestyle factors such as nutrition in successful aging. The current study examined if one year of supplementation with lutein (L) and zeaxanthin (Z), two nutrients with known antioxidative properties and cognitive benefits, impacted structural brain outcomes in older adults using a double-blind, randomized, placebo-controlled trial design. Community-dwelling older adults (20 males and 27 females) aged 65–87 years (M = 71.8 years, SD = 6.04 years) were randomized into supplement (N = 33) and placebo groups (N = 14) using simple randomization. The supplement group received 10 mg L + 2 mg Z daily for 12 months while the placebo group received a visually identical, inert placebo. L and Z were measured via retinal concentrations (macular pigment optical density or MPOD). Structural brain outcomes, focusing on global and frontal-temporal lobe regions, were acquired using both T1-weighted and DTI MRI sequences. We hypothesized that the supplement group would increase, maintain, or show attenuated loss in hypothesized regions-of-interest (ROIs) while the placebo group would show age-related declines in brain structural integrity over the course of the trial. While results showed age-related declines for frontal and temporal gray and white matter volumes, as well as fornix white matter microstructure across both groups, only minimal differences were found between the supplement and placebo groups. However, exploratory analyses showed that individuals who responded better to supplementation (i.e., showed greater increases in MPOD) showed less decline in global and prefrontal gray matter volume than supplement “nonresponders.” While results suggest that one year of L and Z supplementation may have limited effects on structural brain outcomes overall, there may be a subsample of individuals for whom supplementation of L and Z provides greater benefits. ClinicalTrials.gov number, NCT02023645.

Isolation and identification of age-related DNA methylation markers for forensic age-prediction

2014 ◽  
Vol 11 ◽  
pp. 117-125 ◽  
Author(s):  
Shao Hua Yi ◽  
Long Chang Xu ◽  
Kun Mei ◽  
Rong Zhi Yang ◽  
Dai Xin Huang
Keyword(s):  
Dna Methylation ◽  
Isolation And Identification ◽  
Age Related ◽  
Age Prediction

scholarly journals Bayesian Optimisation for Neuroimaging Pre-processing in Brain Age Prediction

2017 ◽  
Author(s):  
Jenessa Lancaster ◽  
Romy Lorenz ◽  
Rob Leech ◽  
James H Cole
Keyword(s):  
Absolute Error ◽  
Processing Parameters ◽  
Training Dataset ◽  
Chronological Age ◽  
Support Vector ◽  
Voxel Size ◽  
Smoothing Kernel ◽  
And Performance ◽  
Brain Age ◽  
Age Prediction

AbstractNeuroimaging-based age predictions using machine learning have been shown to relate to cognitive performance, health outcomes and progression of neurodegenerative disease. However, even leading age-prediction algorithms contain measurement error, motivating efforts to improve experimental pipelines. T1-weighted MRI is commonly used for age prediction, and the pre-processing of these scans involves normalisation to a common template and resampling to a common voxel size, followed by spatial smoothing. Resampling parameters are often selected arbitrarily. Here, we sought to improve brain-age prediction accuracy by optimising resampling parameters using Bayesian optimisation.Using data on N=2001 healthy individuals (aged 16-90 years) we trained support vector machines to i) distinguish between young (<50 years) and old (>50 years) brains and ii) predict chronological age, with accuracy assessed using cross-validation. We also evaluated model generalisability to the Cam-CAN dataset (N=648, aged 18-88 years). Bayesian optimisation was used to identify optimal voxel size and smoothing kernel size for each task. This procedure adaptively samples the parameter space to evaluate accuracy across a range of possible parameters, using independent sub-samples to iteratively assess different parameter combinations to arrive at optimal values.When distinguishing between young and old brains a classification accuracy of 96.25% was achieved, with voxel size = 11.5mm3 and smoothing kernel = 2.3mm. For predicting chronological age, a mean absolute error (MAE) of 5.08 years was achieved, with voxel size = 3.73mm3 and smoothing kernel = 3.68mm. This was compared to performance using default values of 1.5mm3 and 4mm respectively, which gave a MAE = 5.48 years, a 7.3% improvement. When assessing generalisability, best performance was achieved when applying the entire Bayesian optimisation framework to the new dataset, out-performing the parameters optimised for the initial training dataset.Our study demonstrates the proof-of-principle that neuroimaging models for brain age prediction can be improved by using Bayesian optimisation to select more appropriate pre-processing parameters. Our results suggest that different parameters are selected and performance improves when optimisation is conducted in specific contexts. This motivates use of optimisation techniques at many different points during the experimental process, which may result in improved statistical sensitivity and reduce opportunities for experimenter-led bias.

Sign in / Sign up

Export Citation Format

Share Document