We present a Deep Learning framework for the prediction of chronological age from structural magnetic resonance imaging scans. Previous findings associate increased brain age with neurodegenerative diseases and higher mortality rates. However, the importance of brain age prediction goes beyond serving as biomarkers for neurological disorders. Specifically, utilizing convolutional neural network (CNN) analysis to identify brain regions contributing to the prediction can shed light on the complex multivariate process of brain aging. Previous work examined methods to attribute pixel/voxel‐wise contributions to the prediction in a single image, resulting in “explanation maps” that were found noisy and unreliable. To address this problem, we developed an inference scheme for combining these maps across subjects, thus creating a population‐based, rather than a subject‐specific map. We applied this method to a CNN ensemble trained on predicting subjects' age from raw T1 brain images in a lifespan sample of 10,176 subjects. Evaluating the model on an untouched test set resulted in mean absolute error of 3.07 years and a correlation between chronological and predicted age of r = 0.98. Using the inference method, we revealed that cavities containing cerebrospinal fluid, previously found as general atrophy markers, had the highest contribution for age prediction. Comparing maps derived from different models within the ensemble allowed to assess differences and similarities in brain regions utilized by the model. We showed that this method substantially increased the replicability of explanation maps, converged with results from voxel‐based morphometry age studies and highlighted brain regions whose volumetric variability correlated the most with the prediction error.
Background The effect of diet on age-related brain atrophy is largely unproven. Objective To explore the effect of a Mediterranean diet higher in polyphenols and lower in red/processed meat (Green-MED diet) on age-related brain atrophy. Methods This 18-month clinical trial longitudinally measured brain structure volumes by magnetic-resonance-imaging using hippocampal-occupancy (HOC) and lateral-ventricle-volume (LVV) expansion scores as neurodegeneration markers. Abdominally obese/dyslipidemic participants were randomly assigned to (1)-healthy dietary guidelines (HDG), (2)-Mediterranean (MED) diet, or (3)-Green-MED diet (MED diet higher in polyphenols and lower in red/processed meat). All subjects received free gym memberships and physical activity guidance. Both MED groups consumed 28g/day walnuts (+440 mg/d polyphenols). The Green-MED group consumed green-tea (3-4 cups/day) and Mankai (Wolffia-globosa strain, 100g frozen-cubes/day) green shake (+800mg/day polyphenols). Results Among 284 participants (age = 51years; 88% men; BMI = 31.2kg/m2; apolipoprotein E-ε4 genotype = 15.7%), 224 (79%) completed the trial with eligible whole-brain MRIs. The pallidum (-4.2%), third ventricle (+3.9%), and LVV (+2.2%) disclosed the largest volume changes. Compared to younger participants, atrophy was accelerated among those ≥ 50 years [HOC change = -1.0±1.4% vs. -0.06±1.1%; 95% confidence-interval (CI):0.6, 1.3; p<0.001; LVV change = 3.2±4.5% vs. 1.3±4.1%; 95%CI:-3.1, -0.8;p = 0.001]. In subjects ≥50years, HOC decline and LVV expansion were attenuated in both MED groups, with the best outcomes among Green-MED diet participants, as compared to HDG (HOC: -0.8±1.6% vs. -1.3±1.4%;95%CI: -1.5, -0.02;p = 0.042, LVV: 2.3±4.7% vs. 4.3±4.5;95%CI;0.3, 5.2;p = 0.021). Similar patterns were observed among younger subjects. Improved insulin sensitivity over the trial was the strongest parameter associated with brain atrophy attenuation (p<0.05). Greater Mankai, green-tea and walnuts intake and less red and processed meat were significantly and independently associated with reduced HOC decline (p<0.05). Elevated urinary levels of the Mankai-derived polyphenols: urolithin-A (r = 0.24;p = 0.013) and tyrosol (r = 0.26;p = 0.007) were significantly associated with lower HOC decline. Conclusions A Green-MED, high-polyphenol diet, rich in Mankai, green tea and walnuts and low in red/processed meat is potentially neuroprotective for age-related brain atrophy.
The connectome, a comprehensive map of the brain’s anatomical connections, is often summarized as a matrix comprising all dyadic connections among pairs of brain regions. This representation cannot capture higher-order relations within the brain graph. Connectome embedding (CE) addresses this limitation by creating compact vectorized representations of brain nodes capturing their context in the global network topology. Here, nodes “context” is defined as random walks on the brain graph and as such, represents a generative model of diffusive communication around nodes. Applied to group-averaged structural connectivity, CE was previously shown to capture relations between inter-hemispheric homologous brain regions and uncover putative missing edges from the network reconstruction. Here we extend this framework to explore individual differences with a novel embedding alignment approach. We test this approach in two lifespan datasets (NKI: n = 542; Cam-CAN: n = 601) that include diffusion-weighted imaging, resting-state fMRI, demographics and behavioral measures. We demonstrate that modeling functional connectivity with CE substantially improves structural to functional connectivity mapping both at the group and subject level. Furthermore, age-related differences in this structure-function mapping, are preserved and enhanced. Importantly, CE captures individual differences by out-of-sample prediction of age and intelligence. The resulting predictive accuracy was higher compared to using structural connectivity and functional connectivity. We attribute these findings to the capacity of the CE to incorporate aspects of both anatomy (the structural graph) and function (diffusive communication). Our novel approach allows mapping individual differences in the connectome through structure to function and behavior.
It has long been argued that face processing requires disproportionate reliance on holistic processing (HP), relative to that required for nonface object recognition. Nevertheless, whether the holistic nature of face perception is achieved via a unique internal representation or by the employment of an automated attention mechanism is still debated. Previous studies had used the face inversion effect (FIE), a unique face-processing marker, or the face composite task, a gold standard paradigm measuring holistic processing, to examine the validity of these two different hypotheses, with some studies combining the two paradigms. However, the results of such studies remain inconclusive, particularly pertaining to the issue of the two proposed HP mechanisms-an internal representation as opposed to an automated attention mechanism. Here, using the complete composite paradigm design, we aimed to examine whether face rotation yields a nonlinear or a linear drop in HP, thus supporting an account that face processing is based either on an orientation-dependent internal representation or on automated attention. Our results reveal that even a relatively small perturbation in face orientation (30 deg away from upright) already causes a sharp decline in HP. These findings support the face internal representation hypothesis and the notion that the holistic processing of faces is highly orientation-specific.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.