BrainAGE: Revisited and reframed machine learning workflow.

Since the introduction of the BrainAGE method, novel machine learning methods for brain age prediction have continued to emerge. The idea of estimating the chronological age from magnetic resonance images proved to be an interesting field of research due to the relative simplicity of its interpretation and its potential use as a biomarker of brain health. We revised our previous BrainAGE approach, originally utilising relevance vector regression (RVR), and substituted it with Gaussian process regression (GPR), which enables more stable processing of larger datasets, such as the UK Biobank (UKB). In addition, we extended the global BrainAGE approach to regional BrainAGE, providing spatially specific scores for five brain lobes per hemisphere. We tested the performance of the new algorithms under several different conditions and investigated their validity on the ADNI and schizophrenia samples, as well as on a synthetic dataset of neocortical thinning. The results show an improved performance of the reframed global model on the UKB sample with a mean absolute error (MAE) of less than 2 years and a significant difference in BrainAGE between healthy participants and patients with Alzheimer's disease and schizophrenia. Moreover, the workings of the algorithm show meaningful effects for a simulated neocortical atrophy dataset. The regional BrainAGE model performed well on two clinical samples, showing disease-specific patterns for different levels of impairment. The results demonstrate that the new improved algorithms provide reliable and valid brain age estimations.

Approximation of bone mineral density and subcutaneous adiposity using T1-weighted images of the human head.

Bones and brain are intricately connected and scientific interest in their interaction is growing. This has become particularly evident in the framework of clinical applications for various medical conditions, such as obesity and osteoporosis. The adverse effects of obesity on brain health have long been recognised, but few brain imaging studies provide sophisticated body composition measures. Here we propose to extract the following bone- and adiposity-related measures from T1-weighted MR images of the head: an approximation of skull bone mineral density (BMD), skull bone thickness, and two approximations of subcutaneous fat (i.e., the intensity and thickness of soft non-brain head tissue). The measures pertaining to skull BMD, skull bone thickness, and intensi-ty-based adiposity proxy proved to be reliable ( r =.93/.83/.74, p <.001) and valid, with high correlations to DXA-de-rived head BMD values (rho=.70, p <.001) and MRI-derived abdominal subcutaneous adipose volume (rho=.62, p <.001). Thickness-based adiposity proxy had only a low retest reliability ( r =.58, p <.001).The outcomes of this study constitute an important step towards extracting relevant non-brain features from available brain scans.

Low bone mineral density is associated with gray matter volume decrease in UK Biobank.

Objectives: Previous research has found an association of low bone mineral density (BMD) and regional gray matter (GM) volume loss in Alzheimer's disease (AD). We were interested whether BMD is associated with GM volume decrease in brains of a healthy elderly population from the UK Biobank. Materials and methods: T1-weighted images from 5,518 women (MAge = 70.20, SD = 3.54; age range: 65-82 years) and 7,595 men (MAge = 70.84, SD = 3.68; age range: 65-82 years) without neurological or psychiatric impairments were included in voxel-based morphometry (VBM) analysis in CAT12 with threshold-free-cluster-enhancement (TFCE) across the whole brain. Results: We found a significant decrease of GM volume in women in the superior frontal gyri, middle temporal gyri, fusiform gyri, temporal poles, cingulate gyri, precunei, right parahippocampal gyrus and right hippocampus, right ventral diencephalon, and right pre- and postcentral gyrus. Only small effects were found in men in subcallosal area, left basal forebrain and entorhinal area. Conclusion: BMD is associated with low GM volume in women but less in men in regions afflicted in the early-stages of AD even in a sample without neurodegenerative diseases.

Extruded Wheat Bran Consumption Increases Serum Short-Chain Fatty Acids but Does Not Modulate Psychobiological Functions in Healthy Men: A Randomized, Placebo-Controlled Trial.

Background: Incorporation of wheat bran (WB) into food products increases intake of dietary fiber, which has been associated with improved mood and cognition and a lower risk for psychiatric disorders such as depression, with short-chain fatty acids (SCFAs) as candidate mediators of these effects. Modifying WB using extrusion cooking increases SCFA production in vitro relative to unmodified WB. Objective: The aim of this study was to evaluate the effects of extruded WB on psychobiological functioning and the mediating role of SCFAs. Methods: In a randomized, triple-blind, placebo-controlled trial, 69 healthy male participants consumed 55 g of breakfast cereal containing either extruded WB or placebo daily for 28 days. At pre- and post-intervention visits, the cortisol response to experimentally induced stress was measured as a primary outcome. In addition, serum SCFAs and brain-derived neurotrophic factors were quantified as potential mediators. Secondary psychobiological outcomes included subjective stress responses, responses to experimentally induced fear, cortisol awakening response, heart rate variability, and retrospective subjective mood ratings. Intestinal permeability, fecal SCFAs, and stool consistency were measured as secondary biological outcomes. Results: Extruded WB increased serum acetate and butyrate (p < 0.05). None of the primary or secondary outcomes were affected by the intervention. Participants who consumed a placebo exhibited an increase in the percentage of fecal dry weight but did not report increased constipation. Despite these statistically significant effects, these changes were small in magnitude. Conclusions: Extruded WB consumption increased serum short-chain fatty acids but did not modulate psychobiological functions in healthy men. Effective modulation of psychobiological functions may require greater increases in SCFAs than those achieved following extruded WB consumption. Rather than attempting to induce health benefits with a single fiber-rich food, combinations of different fibers, particularly highly fermentable ones, might be needed to further increase SCFA production and uptake in the systemic circulation to observe an effect on psychobiological processes.