Predicting brain age with deep learning from raw imaging data results in\n a reliable and heritable biomarker

2016/12/08 by James H. Cole, Rudra P. K. Poudel, Cole, James H +12 · 13 citations
Environmental Science · Neuroscience · Computer Science · #Health, Environment, Cognitive Aging #Functional Brain Connectivity Studies #Age of Information Optimization

paper · pdf · doi:10.48550/arxiv.1612.02572

Abstract

Machine learning analysis of neuroimaging data can accurately predict\nchronological age in healthy people and deviations from healthy brain ageing\nhave been associated with cognitive impairment and disease. Here we sought to\nfurther establish the credentials of "brain-predicted age" as a biomarker of\nindividual differences in the brain ageing process, using a predictive\nmodelling approach based on deep learning, and specifically convolutional\nneural networks (CNN), and applied to both pre-processed and raw T1-weighted\nMRI data. Firstly, we aimed to demonstrate the accuracy of CNN brain-predicted\nage using a large dataset of healthy adults (N = 2001). Next, we sought to\nestablish the heritability of brain-predicted age using a sample of monozygotic\nand dizygotic female twins (N = 62). Thirdly, we examined the test-retest and\nmulti-centre reliability of brain-predicted age using two samples\n(within-scanner N = 20; between-scanner N = 11). CNN brain-predicted ages were\ngenerated and compared to a Gaussian Process Regression (GPR) approach, on all\ndatasets. Input data were grey matter (GM) or white matter (WM) volumetric maps\ngenerated by Statistical Parametric Mapping (SPM) or raw data. Brain-predicted\nage represents an accurate, highly reliable and genetically-valid phenotype,\nthat has potential to be used as a biomarker of brain ageing. Moreover, age\npredictions can be accurately generated on raw T1-MRI data, substantially\nreducing computation time for novel data, bringing the process closer to giving\nreal-time information on brain health in clinical settings.\n

Citations

• Gradient-based learning applied to document recognition

Cited by

• Explainable Multimodal Regression via Information Decomposition
• MRI-Based Brain Age Estimation with Supervised Contrastive Learning of Continuous Representation
• An Artificial Intelligence Framework for Measuring Human Spine Aging Using MRI
• BrainRotViT: Transformer-ResNet Hybrid for Explainable Modeling of Brain Aging from 3D sMRI
• Genome-wide analysis of brain age identifies 59 associated loci and unveils relationships with mental and physical health
• Disentangling Neurodegeneration with Brain Age Gap Prediction Models: A Graph Signal Processing Perspective
• The BHARAT study: a multi-modal, multi-omics investigation of aging signatures in the Indian population
• Latent Representation Learning from 3D Brain MRI for Interpretable Prediction in Multiple Sclerosis
• Increased cortical thickness and decreased brain age among special operations veterans with blast TBI after a magnesium-ibogaine protocol
• Accelerated Brain Aging Identifies Functional Vulnerability Beyond Chronological Age in Multiple Sclerosis
• NeuroRAD-FM: A Foundation Model for Neuro-Oncology with Distributionally Robust Training
• Building a General SimCLR Self-Supervised Foundation Model Across Neurological Diseases to Advance 3D Brain MRI Diagnoses
• Biomarker Integration and Biosensor Technologies Enabling AI-Driven Insights into Biological Aging

Related

• Predicting brain-age from raw T 1 -weighted Magnetic Resonance Imaging data using 3D Convolutional Neural Networks
• Estimating brain age based on a healthy population with deep learning and structural MRI
• Improving brain age estimates with deep learning leads to identification of novel genetic factors associated with brain aging
• Reframing the Brain Age Prediction Problem to a More Interpretable and Quantitative Approach
• MRI signatures of brain age and disease over the lifespan based on a deep brain network and 14 468 individuals worldwide