Abstract
Machine learning algorithms can be trained to estimate age from brain structural MRI. The difference between an individual’s predicted and chronological age, predicted age difference (PAD), is a phenotype of relevance to aging and brain disease. Here, we present a new deep learning approach to predict brain age from a T1-weighted MRI. The method was trained on a dataset of healthy Icelanders and tested on two datasets, IXI and UK Biobank, utilizing transfer learning to improve accuracy on new sites. A genome-wide association study (GWAS) of PAD in the UK Biobank data (discovery set: N=12378, replication set: N=4456) yielded two sequence variants, rs1452628-T (beta =-0.08, P=1.15times{10}^{-9}) and rs2435204-G (beta =0.102, P=9.73times 1{0}^{-12}). The former is near KCNK2 and correlates with reduced sulcal width, whereas the latter correlates with reduced white matter surface area and tags a well-known inversion at 17q21.31 (H2).
Highlights
Machine learning algorithms can be trained to estimate age from brain structural MRI
The Convolutional neural networks (CNNs) that predicts the test set with the least error is the CNN trained on T1-weighted images followed by the CNN trained on white matter (WM) segmented images (Supplementary Figs. 4 and 5 show scatter plots of the CNN test set predictions against chronological age)
To establish a baseline for the CNN-based techniques, we investigated methods based on feature extraction such as surface-based morphometry (SBM)[26], voxel-based morphometry (VBM)[27], and similarity matrices
Summary
Machine learning algorithms can be trained to estimate age from brain structural MRI. Recent publications, have demonstrated that MRIs can be used to predict chronological age with reasonably good accuracy[1,4,5]. The traditional way to perform brain age prediction is to extract features from brain MRIs followed by classification or regression analysis This includes extracting principal components[4], cortical thickness and surface curvature[6], volume of gray matter (GM), white matter (WM), and cerebrospinal fluid (CSF)[7], and constructing a similarity matrix[8]. Convolutional neural networks (CNNs)[10] are deep learning techniques that are especially powerful for image processing and computer vision They have been applied to brain age prediction[11,12]. ResNets have smoother loss surfaces[25], which in turn helps speeding up convergence. (2) We add inputs to the final CNN layer to factor in information about sex and scanner. (3) Our technique is the first to use deformation information encoded in Jacobian maps to predict brain age. (4) As we have mentioned, our method combines predictions from multiple CNNs by either averaging predictions or by training a data blender
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
