Nicole Sanford

This study investigates how heterogeneity in lifestyle and physical fitness shapes patterns of brain ageing in depression. The study sample included medically healthy adults from the UK Biobank (44-82 years) with a history of depression (n = 896) and non-psychiatric controls (n = 36,206). Heterogeneity through Discriminative Analysis (HYDRA) was used to derive data-driven subgroups within the depression sample based on 224 behavioural and physical health features. Global and voxel-based brain-age-gap-estimates (brainAGE) computed from structural neuroimaging data were used for personalized assessment of brain ageing. Four depression clusters were identified: the "balanced-active" cluster (n = 253) characterized by an active lifestyle and balanced diet; the "low metabolic risk-active" cluster (n = 178) characterized by a health-conscious dietary and exercise lifestyle; the "high metabolic risk-sedentary" cluster (n = 315) who had higher body mass, poor diet, and sedentary behaviour; and the "frailty-moderately active" cluster (n = 150) who had a varied diet but coupled with physical frailty. Mood symptoms were lowest in the "balanced-active" cluster and highest in the "high metabolic risk-sedentary" cluster. While depression overall was associated with minimally accelerated global and local brain ageing in fronto-limbic regions, the "high metabolic risk-sedentary" cluster showed additional, regionally specific acceleration in prefrontal, parahippocampal, and thalamic areas. This work underscores the importance of lifestyle and fitness heterogeneity when examining associations between depression and neurobiological ageing and supports lifestyle-informed stratification as a pathway toward more targeted intervention strategies.

Normative models of brain morphometry quantify individual deviations from typical anatomical patterns and hold promise for enhancing clinical decision-making. However, their clinical utility depends critically on demonstrating generalizability across diverse ethnoracial populations. We previously developed sex-specific, race-neutral normative models for cortical thickness, surface area, and subcortical volumes using brain scans from a large international sample of healthy individuals, as part of the CentileBrain Project, a global initiative to provide open-access, neuroimaging reference models. The primary aim of the present study was to empirically evaluate the generalizability and accuracy of these pretrained models across multiple ethnoracial groups. To this end, we tested model performance in independent samples of healthy individuals from Africa, Asia, Europe, and the Americas, with ethnoracial classification defined either by self-identification or genetic ancestry (N = 4,862). We further compared performance against normative models developed exclusively from a single-population Chinese cohort. Across all groups, as well as in the pooled sample, the pretrained CentileBrain models demonstrated consistently high accuracy, with relative mean absolute error values below 10% for subcortical volume and surface area and below 5% for cortical thickness. Model performance was highly concordant across self-identified and ancestry-defined groups. In a separate analysis, the CentileBrain models performed comparably to a population-specific model when applied to an independent ancestry-matched sample. These findings provide empirical support for the generalizability of race-neutral normative models developed on large and diverse samples and underscore their potential utility for individualized neuroimaging assessment across ethnoracially diverse populations.

Background: This study tested whether lifestyle and fitness features that influence brain health in the general population differentially affect adults with a history of depression. Brain health was assessed using the brain-age-gap-estimate (brainAGE), a personalized index of the brain's biological age. Methods: Medically healthy adults (44-82 years) from the UK Biobank with a history of depression (n=896) or no psychiatric history (n=36,206) were included. Heterogeneity Through Discriminative Analysis was used to cluster the depression group based on 224 lifestyle and fitness features. Global and local (voxel-based) brainAGE were computed from structural neuroimaging data. The study design and implementation involved input from the position of related lived experience. Outcomes: Four depression clusters were identified. The "balanced moderates" cluster (n=253) had good health and balanced lifestyle habits. The "optimal diet and activity" cluster (n=178) had good health, healthy diets, and regular physical activity. The "metabolic risk-sedentary" cluster (n=315) had higher body mass index, poor diet, and sedentary behaviour. The "frailty-low activity" cluster (n=150) had a varied diet coupled with physical frailty. Mood symptoms were lowest in the "balanced moderates" cluster and highest in the "metabolic risk-sedentary" cluster. The presence of a history of depression was associated with older global brainAGE and with local brainAGE in ventromedial prefrontal regions regardless of cluster assignment. The "metabolic risk-sedentary" cluster also exhibited elevated local brainAGE in the hippocampal complex and thalamus. Interpretation: This study highlights the heterogeneity in lifestyle and fitness factors among adults with a history of depression, underscoring the detrimental influence of depression as well as poor diet and physical inactivity on the biological age of the brain.

Structural neuroimaging data have been used to compute an estimate of the biological age of the brain (brain‐age) which has been associated with other biologically and behaviorally meaningful measures of brain development and aging. The ongoing research interest in brain‐age has highlighted the need for robust and publicly available brain‐age models pre‐trained on data from large samples of healthy individuals. To address this need we have previously released a developmental brain‐age model. Here we expand this work to develop, empirically validate, and disseminate a pre‐trained brain‐age model to cover most of the human lifespan. To achieve this, we selected the best‐performing model after systematically examining the impact of seven site harmonization strategies, age range, and sample size on brain‐age prediction in a discovery sample of brain morphometric measures from 35,683 healthy individuals (age range: 5–90 years; 53.59% female). The pre‐trained models were tested for cross‐dataset generalizability in an independent sample comprising 2101 healthy individuals (age range: 8–80 years; 55.35% female) and for longitudinal consistency in a further sample comprising 377 healthy individuals (age range: 9–25 years; 49.87% female). This empirical examination yielded the following findings: (1) the accuracy of age prediction from morphometry data was higher when no site harmonization was applied; (2) dividing the discovery sample into two age‐bins (5–40 and 40–90 years) provided a better balance between model accuracy and explained age variance than other alternatives; (3) model accuracy for brain‐age prediction plateaued at a sample size exceeding 1600 participants. These findings have been incorporated into CentileBrain (https://centilebrain.org/#/brainAGE2), an open‐science, web‐based platform for individualized neuroimaging metrics.

The brain‐age‐gap estimate (brainAGE) quantifies the difference between chronological age and age predicted by applying machine‐learning models to neuroimaging data and is considered a biomarker of brain health. Understanding sex differences in brainAGE is a significant step toward precision medicine. Global and local brainAGE (G‐brainAGE and L‐brainAGE, respectively) were computed by applying machine learning algorithms to brain structural magnetic resonance imaging data from 1113 healthy young adults (54.45% females; age range: 22–37 years) participating in the Human Connectome Project. Sex differences were determined in G‐brainAGE and L‐brainAGE. Random forest regression was used to determine sex‐specific associations between G‐brainAGE and non‐imaging measures pertaining to sociodemographic characteristics and mental, physical, and cognitive functions. L‐brainAGE showed sex‐specific differences; in females, compared to males, L‐brainAGE was higher in the cerebellum and brainstem and lower in the prefrontal cortex and insula. Although sex differences in G‐brainAGE were minimal, associations between G‐brainAGE and non‐imaging measures differed between sexes with the exception of poor sleep quality, which was common to both. While univariate relationships were small, the most important predictor of higher G‐brainAGE was self‐identification as non‐white in males and systolic blood pressure in females. The results demonstrate the value of applying sex‐specific analyses and machine learning methods to advance our understanding of sex‐related differences in factors that influence the rate of brain aging and provide a foundation for targeted interventions.

Background: The brain-age-gap estimate (brainAGE) quantifies the difference between chronological age and age predicted by applying machine-learning models to neuroimaging data, and is considered a biomarker of brain health. Understanding sex-differences in brainAGE is a significant step toward precision medicine. Methods: Global and local brainAGE (G-brainAGE and L-brainAGE, respectively) were computed by applying machine learning algorithms to brain structural magnetic resonance imaging data from 1113 healthy young adults (54.45% females; age range: 22-37 years) participating in the Human Connectome Project. Sex-differences were determined in G-brainAGE and L-brainAGE. Random forest regression was used to determine sex-specific associations between G-brainAGE and non-imaging measures pertaining to sociodemographic characteristics and mental, physical, and cognitive functions. Results: L-brainAGE showed sex-specific differences in brain ageing. In females, compared to males, L-brainAGE was higher in the cerebellum and brainstem and lower in the prefrontal cortex and insula. Although sex-differences in G-brainAGE were minimal, associations between G-brainAGE and non-imaging measures differed between sexes with the exception for poor sleep quality, which was common to both. The most important predictor of higher G-brainAGE was non-white race in males and systolic blood pressure in females. Conclusions: The results demonstrate the value of applying sex-specific analyses and machine learning methods to advance our understanding of sex-related differences in factors that influence the rate of brain ageing and provide a foundation for targeted interventions.

While advances in the field of functional magnetic resonance imaging (fMRI) provide new opportunities to study brain networks underlying the experience of hallucinations in psychosis, there are methodological challenges unique to symptom-capture studies. We extracted brain networks activated during hallucination-capture for schizophrenia patients when fMRI data collected from two sites was merged (combined N = 27). A multidimensional analysis technique was applied, which would allow separation of brain networks involved in the hallucinatory experience itself from those involved in the motor response of indicating the beginning and end of the perceived hallucinatory experience. To avoid reverse inference when attributing a function (e.g., a hallucination) to anatomical regions, it was required that longer hallucinatory experiences produce extended brain responses relative to shorter. For radio-speech sound files, an auditory perception brain network emerged, and displayed speech-duration-dependent hemodynamic responses (HDRs). However, in the hallucination-capture blocks, no network showed hallucination-duration-dependent HDRs, but a retrieved network that was anatomically classified as motor response emerged. During symptom capture of hallucinations during fMRI, no HDR showed duration dependence, but a brain network anatomically matching the motor response network was retrieved. Previous reports on brain networks detected by fMRI during hallucination capture are reviewed in this context; namely, that the brain networks interpreted as involved in hallucinations may in fact be involved only in the motor response indicating the onset of the hallucination.

Background: Working memory (WM) impairment in schizophrenia substantially impacts functional outcome. Although the dorsolateral pFC has been implicated in such impairment, a more comprehensive examination of brain networks comprising pFC is warranted. The present research used a whole-brain, multi-experiment analysis to delineate task-related networks comprising pFC. Activity was examined in schizophrenia patients across a variety of cognitive demands. Methods: One hundred schizophrenia patients and 102 healthy controls completed one of four fMRI tasks: a Sternberg verbal WM task, a visuospatial WM task, a Stroop set-switching task, and a thought generation task (TGT). Task-related networks were identified using multi-experiment constrained PCA for fMRI. Effects of task conditions and group differences were examined using mixed-model ANOVA on the task-related time series. Correlations between task performance and network engagement were also performed. Results: Four spatially and temporally distinct networks with pFC activation emerged and were postulated to subserve (1) internal attention, (2) auditory–motor attention, (3) motor responses, and (4) task energizing. The “energizing” network—engaged during WM encoding and diminished in patients—exhibited consistent trend relationships with WM capacity across different data sets. The dorsolateral-prefrontal-cortex-dominated “internal attention” network exhibited some evidence of hypoactivity in patients, but was not correlated with WM performance. Conclusions: Multi-experiment analysis allowed delineation of task-related, pFC-anchored networks across different cognitive constructs. Given the results with respect to the early-responding “energizing” network, WM deficits in schizophrenia may arise from disruption in the “energization” process described by Donald Stuss' model of pFC functions.