Older Brain Age Research Articles

Diabetes, Prediabetes, and Brain Aging: The Role of Healthy Lifestyle.

Diabetes is a well-known risk factor for dementia. We investigated the association between (pre)diabetes and older brain age and whether this can be attenuated by modifiable lifestyle behaviors. The study included 31,229 dementia-free adults from the UK Biobank between the ages of 40 and 70 years. Glycemic status (normoglycemia, prediabetes, or diabetes) was ascertained based on medical history, medication use, and HbA1c measured at baseline. Information on cardiometabolic risk factors (obesity, hypertension, low HDL, and high triglycerides) and lifestyle behaviors (smoking, drinking, and physical activity) was also collected at baseline. Participants underwent up to two brain MRI scans over 11 years of follow-up. Brain age was estimated using a machine learning model based on 1,079 brain MRI phenotypes and used to calculate brain age gap (BAG; i.e., brain age minus chronological age). At baseline, 13,518 participants (43.3%) had prediabetes and 1,149 (3.7%) had diabetes. Prediabetes (β = 0.22 [95% CI 0.10, 0.34]) and diabetes (2.01 [1.70, 2.32]) were both associated with significantly higher BAG, and diabetes was further associated with significant increase in BAG over time (0.27 [0.01, 0.53]). The association between (pre)diabetes and higher BAG was more pronounced in men and in people with two or more cardiometabolic risk factors. In joint exposure analysis, having a healthy lifestyle (i.e., no smoking, no heavy drinking, and high physical activity) significantly attenuated the diabetes-BAG association. Diabetes and even prediabetes are associated with accelerated brain aging, especially among men and people with poor cardiometabolic health. However, a healthy lifestyle may counteract this.

Sex-Dependent Effects of Cardiometabolic Health and APOE4 on Brain Age: A Longitudinal Cohort Study.

The aging population is growing faster than all other demographic strata. With older age comes a greater risk of health conditions such as obesity and high blood pressure (BP). These cardiometabolic risk factors (CMRs) exhibit prominent sex differences in midlife and aging, yet their influence on brain health in females vs males is largely unexplored. In this study, we investigated sex differences in relationships between BP, body mass index (BMI), and brain age over time and tested for interactions with APOE ε4 genotype (APOE4), a known genetic risk factor of Alzheimer disease. The sample included participants from 2 United Kingdom-based longitudinal birth cohorts, the Lothian Birth Cohort (1936) and Insight 46 (1946). Participants with MRI data from at least 1 time point were included to evaluate sex differences in associations between CMRs and brain age. The open-access software package brainageR 2.1 was used to estimate brain age for each participant. Linear mixed-effects models were used to assess the relationships between brain age, BMI, BP, and APOE4 status (i.e., carrier vs noncarrier) in males and females over time. The combined sample comprised 1,120 participants (48% female) with a mean age (SD) of 73 (0.72) years in the Lothian Birth Cohort and 71 (0.68) years in Insight 46 at the time point 1 assessment. Approximately 30% of participants were APOE4 carriers. Higher systolic and diastolic BP was significantly associated with older brain age in females only (β = 0.43-0.56, p < 0.05). Among males, higher BMI was associated with older brain age across time points and APOE4 groups (β = 0.72-0.77, p < 0.05). In females, higher BMI was linked to older brain age among APOE4 noncarriers (β = 0.68-0.99, p < 0.05), whereas higher BMI was linked to younger brain age among carriers, particularly at the last time point (β = -1.75, p < 0.05). This study indicates sex-dependent and time-dependent relationships between CMRs, APOE4 status, and brain age. Our findings highlight the necessity of sex-stratified analyses to elucidate the role of CMRs in individual aging trajectories, providing a basis for developing personalized preventive interventions.

Association of Cardiovascular Health With Brain Age Estimated Using Machine Learning Methods in Middle-Aged and Older Adults.

Cardiovascular health (CVH) has been associated with cognitive decline and dementia, but the extent to which CVH affects brain health remains unclear. We investigated the association of CVH, assessed using Life's Essential 8 (LE8), with neuroimaging-based brain age and brain-predicted age difference (brain-PAD). This longitudinal community-based study was based on UK Biobank participants aged 40-69 years who were free from dementia and other neurologic diseases at baseline. LE8 score at baseline was assessed with 8 measures and tertiled as low, moderate, and high CVH. Structural and functional brain MRI scans were performed approximately 9 years after baseline, and 1,079 measures from 6 neuroimaging modalities were used to model brain age. A Least Absolute Shrinkage and Selection Operator regression model was trained in 4,355 healthy participants and then used to calculate brain age and brain-PAD in the whole population. Data were analyzed using linear regression models. The study included 32,646 participants (mean age at baseline 54.74 years; 53.44% female; mean LE8 score: 71.90). In multivariable-adjusted linear regression, higher LE8 score was associated with younger brain age (β [95% CI] -0.037 [-0.043 to -0.031]) and more negative brain-PAD (β [95% CI] -0.043 [-0.048 to -0.038]) (brain looks younger for chronological age). Compared with high CVH, low/moderate CVH was associated with older brain age (β [95% CI] 1.030 [0.852-1.208]/0.475 [0.303-0.647]) and increased brain-PAD (β [95% CI] 1.193 [1.029-1.357]/0.528 [0.370-0.686]). The associations between low CVH and older brain age/brain-PAD remained similar and significant in both middle-aged (β [95% CI] 1.199 [0.992-1.405]/1.351 [1.159-1.542]) and older adults (β [95% CI] 0.764 [0.417-1.110]/0.948 [0.632-1.263]). Low CVH is associated with older brain age and greater brain-PAD, even among middle-aged adults. Our findings suggest that optimizing CVH could support brain health. The main limitation of our study is that the study sample was healthier than the general population, thus caution is required when generalizing our findings to other populations.

Association between Resting Heart Rate and Machine Learning-Based Brain Age in Middle- and Older-Age

BackgroundResting heart rate (RHR), has been related to increased risk of dementia, but the relationship between RHR and brain age is unclear.ObjectiveWe aimed to investigate the association of RHR with brain age and brain age gap (BAG, the difference between predicted brain age and chronological age) assessed by multimodal Magnetic Resonance Imaging (MRI) in mid- and old-aged adults.DesignA longitudinal study from the UK Biobank neuroimaging project where participants underwent brain MRI scans 9+ years after baseline.SettingA population-based study.ParticipantsA total of 33,381 individuals (mean age 54.74 ± 7.49 years; 53.44% female).MeasurementsBaseline RHR was assessed by blood pressure monitor and categorized as <60, 60–69 (reference), 70–79, or ≥80 beats per minute (bpm). Brain age was predicted using LASSO through 1,079 phenotypes in six MRI modalities (including T1-weighted MRI, T2-FLAIR, T2*, diffusion-MRI, task fMRI, and resting-state fMRI). Data were analyzed using linear regression models.ResultsAs a continuous variable, higher RHR was associated with older brain age (β for per 1-SD increase: 0.331, 95% [95% confidence interval, CI]: 0.265, 0.398) and larger BAG (β: 0.263, 95% CI: 0.202, 0.324). As a categorical variable, RHR 70–79 bpm and RHR ≥80 bpm were associated with older brain age (β [95% CI]: 0.361 [0.196, 0.526] / 0.737 [0.517, 0.957]) and larger BAG (0.256 [0.105, 0.407] / 0.638 [0.436, 0.839]), but RHR< 60 bpm with younger brain age (−0.324 [−0.500, −0.147]) and smaller BAG (−0.230 [−0.392, −0.067]), compared to the reference group. These associations between elevated RHR and brain age were similar in both middle-aged (<60) and older (≥60) adults, whereas the association of RHR< 60 bpm with younger brain age and larger BAG was only significant among middle-aged adults. In stratification analysis, the association between RHR ≥80 bpm and older brain age was present in people with and without CVDs, while the relation of RHR 70–79 bpm to brain age present only in people with CVD.ConclusionHigher RHR (>80 bpm) is associated with older brain age, even among middle-aged adults, but RHR< 60 bpm is associated with younger brain age. Greater RHR could be an indicator for accelerated brain aging.

Association of resting heart rate with cognitive decline and brain age in older adults: a large‐scale cohort study in the UK Biobank

AbstractBackgroundThe relationship of resting heart rate (RHR), a common indicator of cardiovascular health, with cognitive decline is unclear. We aimed to investigate the association of RHR with cognitive trajectory and brain age assessed by brain Magnetic Resonance Imaging (MRI) in the UK Biobank.MethodWithin the UK Biobank 18,210 chronic brain disorders‐free participants aged 60+ were followed for up to 16 years. RHR was assessed by blood pressure monitor and categorized into &lt;60, 60‐69 (reference), 70‐79, and ≥80 beats per minute (bpm). Cognitive function including global cognition, numeric memory, prospective memory, pairs matching, fluid intelligence, and reaction time was collected at baseline and each follow‐up wave. A subsample of 10,458 participants underwent brain MRI (T1‐weighted MRI, T2‐FLAIR, T2∗, diffusion‐MRI, task fMRI, and resting‐state fMRI) to detect brain age and brain predicted age difference (PAD) using Lasso regression. Data were analyzed using mixed‐effects model and linear regression model.ResultCompared to reference group, RHR≥80 bpm was associated with faster decline in global cognition (β: ‐0.03, 95% confidence interval [CI]: ‐0.05, ‐0.01) and reaction time (β: ‐0.03, 95% CI: ‐0.07, ‐0.01). In the MRI data analysis, RHR≥80 bpm was associated with older brain age (β [95% CI]: 0.50 [0.25, 0.76]) and bigger brain PAD (β [95% CI]: 0.50 [0.26, 0.74]).ConclusionHigher RHR is associated with faster cognitive decline and older brain age. Monitoring and controlling elevated RHR could be a promising interventional target in preserving cognitive function.

Association between cardiovascular health and brain age: a large population‐based study in the UK Biobank

AbstractBackgroundCardiovascular health (CVH) has been associated with cognitive decline and dementia, but the extent to which CVH impact brain age remains unclear. We aimed to investigate the association of CVH assessed by Life’s Essential 8 (LE8) score with neuroimaging‐based brain age and brain predicted age difference (PAD).MethodThe study included a total of 34098 participants (mean age 54.82±7.50 years; 53.21% female) from the UK Biobank, who underwent brain magnetic resonance imaging (MRI) scans 9 years after baseline. A brain age prediction model with the best prediction performance was trained from 9 candidate models (Lasso, support vector regression, and XGBoost, with or without the addition of 2 variable selection methods, respectively) based on T1‐weighted MRI, T2‐FLAIR, T2∗, diffusion‐MRI, task fMRI, and resting‐state fMRI in 4355 healthy participants. Using the model, we calculated brain age and brain PAD (predicted age‐chronological age). Baseline LE8 score was calculated by the mean score of 8 metrics (diet, physical activity, nicotine exposure, sleep health, body mass index, blood lipids, blood glucose, and blood pressure) and tertiled as low, moderate, and high CVH. Data were analyzed using linear regression models.ResultAt baseline, LE8 score was ranged from 15.00 to 100.00. In linear regression adjusted for age, sex and education, better CVH was associated with younger brain age (β [95% CI]: ‐0.029 [‐0.032, ‐0.026]) and smaller brain PAD (β [95% CI]: ‐0.033 [‐0.035, ‐0.030]). Compared to moderate CVH, low CVH was associated with older brain age (β [95% CI]: 0.444 [0.345, 0.543]) and bigger brain PAD (β [95% CI]: 0.513 [0.422, 0.604]), whereas high CVH was associated with younger brain age (β [95% CI]: ‐0.360 [‐0.459, ‐0.261]) and smaller brain PAD (β [95% CI]: ‐0.398 [‐0.489, ‐0.307]). The additional adjustment did not alter the relationships observed. After stratification by age, cardiovascular disease (CVD), or APOE ε4, the association remained significant among adults aged 40‐60 years and 60+ years, among CVD and non‐CVD, as well as among APOE ε4 carriers and non‐ carriers.ConclusionBetter cardiovascular health is associated with younger brain age and reduced brain PAD. This association is significant even in middle‐aged, non‐CVD and non‐carriers adults.

BrainAGE as a measure of maturation during early adolescence

Abstract The Brain-Age Gap Estimation (BrainAGE) is an important new tool that purports to evaluate brain maturity when used in adolescent populations. However, it is unclear whether BrainAGE tracks with other maturational metrics in adolescence. In the current study, we related BrainAGE to metrics of pubertal and cognitive development using both a previously validated model and a novel model trained specifically on an early adolescent population. The previously validated model was used to predict BrainAGE in two age bands, 9-11 and 10-13 years old, while the novel model was used with 9-11 year olds only. Across both models and age bands, an older BrainAGE was related to more advanced pubertal development. The relationship between BrainAGE and cognition was less clear, with conflicting relationships across the two models. Additionally, longitudinal analysis revealed moderate to high stability in BrainAGE across early adolescence. The results of the current study provide initial evidence that BrainAGE tracks with some metrics of maturation, including pubertal development. However, the conflicting results between BrainAGE and cognition lead us to question the utility of these models for non-biological processes.

Social isolation from childhood to mid-adulthood: is there an association with older brain age?

Older brain age - as estimated from structural MRI data - is known to be associated with detrimental mental and physical health outcomes in older adults. Social isolation, which has similar detrimental effects on health, may be associated with accelerated brain aging though little is known about how different trajectories of social isolation across the life course moderate this association. We examined the associations between social isolation trajectories from age 5 to age 38 and brain age assessed at age 45. We previously created a typology of social isolation based on onset during the life course and persistence into adulthood, using group-based trajectory analysis of longitudinal data from a New Zealand birth cohort. The typology comprises four groups: 'never-isolated', 'adult-only', 'child-only', and persistent 'child-adult' isolation. A brain age gap estimate (brainAGE) - the difference between predicted age from structural MRI date and chronological age - was derived at age 45. We undertook analyses of brainAGE with trajectory group as the predictor, adjusting for sex, family socio-economic status, and a range of familial and child-behavioral factors. Older brain age in mid-adulthood was associated with trajectories of social isolation after adjustment for family and child confounders, particularly for the 'adult-only' group compared to the 'never-isolated' group. Although our findings are associational, they indicate that preventing social isolation, particularly in mid-adulthood, may help to avert accelerated brain aging associated with negative health outcomes later in life.

Associations of Enlarged Perivascular Spaces With Brain Lesions, Brain Age, and Clinical Outcomes in Chronic Traumatic Brain Injury.

Enlarged perivascular spaces (ePVS) have been identified as a key signature of glymphatic system dysfunction in neurologic conditions. The incidence and clinical implications of ePVS after traumatic brain injury (TBI) are not yet understood. We investigated whether individuals with chronic moderate-to-severe TBI had an increased burden of ePVS and whether ePVS burden is modulated by the presence of focal lesions, older brain age, and poorer sleep quality. We examined whether an increased burden of ePVS was associated with poorer cognitive and emotional outcomes. Using a cross-sectional design, participants with a single moderate-to-severe chronic TBI (sustained ≥10 years ago) were recruited from an inpatient rehabilitation program. Control participants were recruited from the community. Participants underwent 3T brain MRI, neuropsychological assessment, and clinical evaluations. ePVS burden in white matter was quantified using automated segmentation. The relationship between the number of ePVS, group membership, focal lesions, brain age, current sleep quality, and outcome was modeled using negative binomial and linear regressions. This study included 100 participants with TBI (70% male; mean age = 56.8 years) and 75 control participants (54.3% male; mean age = 59.8 years). The TBI group had a significantly greater burden of ePVS (prevalence ratio rate [PRR] = 1.29, p = 0.013, 95% CI 1.05-1.57). The presence of bilateral lesions was associated with greater ePVS burden (PRR = 1.41, p = 0.021, 95% CI 1.05-1.90). There was no association between ePVS burden, sleep quality (PRR = 1.01, p = 0.491, 95% CI 0.98-1.048), and sleep duration (PRR = 1.03, p = 0.556, 95% CI 0.92-1.16). ePVS was associated with verbal memory (β = -0.42, p = 0.006, 95% CI -0.72 to -0.12), but not with other cognitive domains. The burden of ePVS was not associated with emotional distress (β = -0.70, p = 0.461, 95% CI -2.57 to 1.17) or brain age (PRR = 1.00, p = 0.665, 95% CI 0.99-1.02). TBI is associated with a greater burden of ePVS, especially when there have been bilateral brain lesions. ePVS was associated with reduced verbal memory performance. ePVS may indicate ongoing impairments in glymphatic system function in the chronic postinjury period.

Associations Between Thinner Retinal Neuronal Layers and Suboptimal Brain Structural Integrity in a Middle-Aged Cohort.

The retina has potential as a biomarker of brain health and Alzheimer's disease (AD) because it is the only part of the central nervous system which can be easily imaged and has advantages over brain imaging technologies. Few studies have compared retinal and brain measurements in a middle-aged sample. The objective of our study was to investigate whether retinal neuronal measurements were associated with structural brain measurements in a middle-aged population-based cohort. Participants were members of the Dunedin Multidisciplinary Health and Development Study (n=1037; a longitudinal cohort followed from birth and at ages 3, 5, 7, 9, 11, 13, 15, 18, 21, 26, 32, 38, and most recently at age 45, when 94% of the living Study members participated). Retinal nerve fibre layer (RNFL) and ganglion cell-inner plexiform layer (GC-IPL) thickness were measured by optical coherence tomography (OCT). Brain age gap estimate (brainAGE), cortical surface area, cortical thickness, subcortical grey matter volumes, white matter hyperintensities, were measured by magnetic resonance imaging (MRI). Participants with both MRI and OCT data were included in the analysis (RNFL n=828, female n=413 [49.9%], male n=415 [50.1%]; GC-IPL n=825, female n=413 [50.1%], male n=412 [49.9%]). Thinner retinal neuronal layers were associated with older brain age, smaller cortical surface area, thinner average cortex, smaller subcortical grey matter volumes, and increased volume of white matter hyperintensities. These findings provide evidence that the retinal neuronal layers reflect differences in midlife structural brain integrity consistent with increased risk for later AD, supporting the proposition that the retina may be an early biomarker of brain health.

Alzheimer's disease risk biomarkers: progress and challenges

Alzheimer's disease risk biomarkers: progress and challenges

MRI-derived brain age as a biomarker of ageing in rats: validation using a healthy lifestyle intervention

The difference between brain age predicted from MRI and chronological age (the so-called BrainAGE) has been proposed as an ageing biomarker. We analyse its cross-species potential by testing it on rats undergoing an ageing modulation intervention. Our rat brain age prediction model combined Gaussian process regression with a classifier and achieved a mean absolute error (MAE) of 4.87 weeks using cross-validation on a longitudinal dataset of 31 normal ageing rats. It was then tested on two groups of 24 rats (MAE = 9.89 weeks, correlation coefficient = 0.86): controls vs. a group under long-term environmental enrichment and dietary restriction (EEDR). Using a linear mixed-effects model, BrainAGE was found to increase more slowly with chronological age in EEDR rats (p=0.015 for the interaction term). Cox regression showed that older BrainAGE at 5 months was associated with higher mortality risk (p=0.03). Our findings suggest that lifestyle-related prevention approaches may help to slow down brain ageing in rodents and the potential of BrainAGE as a predictor of age-related health outcomes.

Association of childhood lead exposure with MRI measurements of structural brain integrity in midlife

BACKGROUND AND AIM: Childhood lead exposure has been linked to disrupted brain development, but long-term consequences for structural brain integrity are unknown. This study tested the hypothesis that childhood lead exposure is associated with magnetic resonance imaging (MRI) measurements of lower structural integrity of the brain in midlife. METHODS: Hypotheses tested in the Dunedin Study, a New Zealand-based population-representative 1972-1973 birth cohort (N=564 analytic sample) followed to age 45 years. Childhood blood-lead levels measured at age 11 years were combined with MRI-measures of brain integrity at age 45 years, including gray matter (cortical thickness, surface area, hippocampal volume), white matter (white matter hyperintensities, fractional anisotropy), and brainAGE, a composite index of the gap between chronological age and a machine-learning algorithm-estimated brain age. Age-45 cognitive function was also assessed, objectively via the Wechsler Adult Intelligence Scale–IV and subjectively via informant and self-reports. RESULTS:Mean(SD) age-11 blood-lead level was 10.99(4.63)µg/dL. After adjusting for covariates, each 5µg/dL higher childhood blood-lead level was significantly associated with 1.19-cm2 smaller cortical surface area (95%CI:-2.35,-0.02, P=.05), 0.10-cm3 smaller hippocampal volume (95%CI:-0.17,-0.03, P=.006), lower global fractional anisotropy (b=-0.12, 95%CI:-0.24,-0.01, P=.04), and 0.77-years older brainAGE (95%CI:0.02, 1.51, P=.05). There were no statistically significant associations between blood-lead level and log-transformed white matter hyperintensity volume (b=0.05 log mm3, 95%CI:-0.02, 0.13, P=.17) or mean cortical thickness (b=-0.004 mm, 95%CI:-0.012, 0.004, P=.39). Each 5µg/dL higher childhood blood-lead level was significantly associated with a 2.07-point lower score (95%CI:-3.39,-0.74, P=.002) in age-45 IQ, and a 0.12-point higher score (95%CI:0.01, 0.23, P=.03) on informant-rated cognitive problems. There was no statistically significant association between childhood blood-lead levels and self-reported cognitive problems (b=-0.02 points, 95%CI: -0.10, 0.07, P=.68). CONCLUSIONS:In this four-decade study, higher childhood blood-lead level was associated with lower structural brain integrity in midlife. Childhood lead exposure may have long-term consequences for adult brain health. KEYWORDS: Heavy metals, Neurodegenerative outcomes, Environmental epidemiology

Neocortical Age and Fluid Ability: Greater Accelerated Brain Aging for Thickness, but Smaller for Surface Area, in High Cognitive Ability Individuals

Biological (BA) and chronological (CA) age may or may not fit. The available evidence reveals remarkable individual differences in the overlap/mismatch between BA and CA. Increased mismatch can be interpreted as delayed (BA/CA < 1) or accelerated biological aging (BA/CA > 1). Body and brain health are correlated and both predict aging outcomes associated with physical and mental fitness. Moreover, research has shown that older brain age at midlife correlates negatively with cognitive ability measured in early childhood, which suggests early life predisposition to accelerated aging in adulthood. Under this framework, here we test if increased cognitive ability is associated with delayed brain aging, analyzing structural MRI data of 188 individuals, sixty of whom were recruited from MENSA, an association comprising individuals who obtained cognitive ability scores in the top 2 percent of the population. These high ability individuals (HCA) showed an average advantage of 33 IQ points, on a fluid reasoning test they completed for this research, over those other recruited because of their average cognitive ability (ACA). Next, brain age was computed at the individual level for two distinguishable neocortical features (thickness and surface area) according to models trained in an independent large-scale sample of 2377 individuals. Results revealed a stronger pattern of accelerated brain aging in HCA compared to ACA individuals for thickness, while the opposite pattern was suggested for surface area. The findings align well with the greater relevance of individual differences in cortical surface area for enhancing our understanding of cognitive differences at the brain level.

Associations Between a New Biomarker of Elevated Chronic Inflammation and Accelerated Aging

To further understand and measure the association between chronic inflammation, aging, and age-related diseases, broadly applicable standard biomarkers of systemic chronic inflammation are needed. We tested whether elevated blood levels of the emerging chronic inflammation marker soluble urokinase plasminogen activator receptor (suPAR) were associated with accelerated aging, lower functional capacity, and cognitive decline. We used data from the population-representative longitudinal Dunedin Study (N=875). Plasma suPAR levels were analyzed at ages 38 and 45 years. We performed regression analyses adjusted for sex, smoking, and C-reactive protein. suPAR levels increased from 2.39 ng/mL (SD 0.89) at age 38 to 3.01 (SD 1.03) at age 45 years. Elevated suPAR was associated with accelerated pace of biological aging across multiple organ systems (β 0.28, 95% CI 0.21–0.35), older facial appearance (β 0.16, 95% CI 0.10–0.22), and with structural signs of older brain age (β 0.06, 95% CI -0.00–0.13). Moreover, participants with higher suPAR levels had lower functional capacity (more physical limitations [β 0.24, 95% CI 0.18–0.30]; slower gait speed [β -0.14, 95% CI -0.20; -0.08]) and greater decline in cognitive function (β -0.07, 95% CI -0.13; -0.01) from childhood to adulthood compared to those with lower suPAR levels. Finally, improvements in health habits between age 38 and 45 (smoking cessation or increased physical activity) were associated with less steep increases in suPAR levels over those years. Our findings provide initial support for the utility of suPAR in studying the role of chronic inflammation in accelerated aging and functional decline.

Brain-age in midlife is associated with accelerated biological aging and cognitive decline in a longitudinal birth cohort

An individual’s brainAGE is the difference between chronological age and age predicted from machine-learning models of brain-imaging data. BrainAGE has been proposed as a biomarker of age-related deterioration of the brain. Having an older brainAGE has been linked to Alzheimer’s, dementia, and mortality. However, these findings are largely based on cross-sectional associations which can confuse age differences with cohort differences. To illuminate the validity of brainAGE as a biomarker of accelerated brain aging, a study is needed of a large cohort all born in the same year who nevertheless vary on brainAGE. In the Dunedin Study, a population-representative 1972–73 birth cohort, we measured brainAGE at age 45 years, as well as the pace of biological aging and cognitive decline in longitudinal data from childhood to midlife (N = 869). In this cohort, all chronological age 45 years, brainAGE was measured reliably (ICC = 0.81) and ranged from 24 to 72 years. Those with older midlife brainAGEs tended to have poorer cognitive function in both adulthood and childhood, as well as impaired brain health at age 3. Furthermore, those with older brainAGEs had an accelerated pace of biological aging, older facial appearance, and early signs of cognitive decline from childhood to midlife. These findings help to validate brainAGE as a potential surrogate biomarker for midlife intervention studies that seek to measure dementia-prevention efforts in midlife. However, the findings also caution against the assumption that brainAGE scores represent only age-related deterioration of the brain as they may also index central nervous system variation present since childhood.

Premature white matter aging in patients with right mesial temporal lobe epilepsy: A machine learning approach based on diffusion MRI data.

Brain age prediction based on machine learning has been applied to various neurological diseases to discover its clinical values. By this innovative approach, it has been reported that the patients with refractory epilepsy had premature brain aging. Of refractory epilepsy, right and left subtypes of mesial temporal lobe epilepsy (MTLE) are the most common forms and exhibit distinct patterns in white matter alterations. So far, it is unclear whether these two subtypes of MTLE would have difference in white matter aging due to distinct white matter alterations. To address this issue, a machine learning based brain age model using diffusion MRI data was established to investigate biological age of white matter tracts. All diffusion MRI datasets were obtained from the same 3-Tesla MRI scanner. To build the brain age prediction model, diffusion MRI datasets of 300 healthy participants were processed to extract age-relevant diffusion indices from 76 major white matter tracts. The extracted diffusion indices underwent Gaussian process regression to build the prediction model for white matter brain age. The model was validated in an independent testing set (N = 40) to ensure no overfitting of the model. The model was then applied to patients with right and left MTLE and matched controls (right MTLE: N = 17, left MTLE: N = 18, controls: N = 37), and predicted age difference (PAD) was obtained by calculating the difference between each individual's predicted brain age and chronological age. The higher PAD score indicated older brain age. The results showed that right MTLE exhibited older predicted brain age than the other two groups (PAD of right MTLE=10.9 years [p<0.05 against left MTLE; p<0.001 against control]; PAD of left MTLE=2.2 years [p>0.1 against control]; PAD of controls=0.82 years). Patients with right and left MTLE showed strong correlations of the PAD scores with age of onset and duration of illness, but both groups showed opposite directions of correlations. In right MTLE, positive correlation of PAD with seizure frequency was found, and the right uncinate fasciculus was the most attributable tract to the increase in PAD. In conclusion, the present study found that patients with right MTLE exhibited premature white matter brain aging and their PAD scores were correlated with seizure frequency. Therefore, PAD is a potentially useful indicator of white matter impairment and disease severity in patients with right MTLE.

S141. Predicting brain age from the electroencephalogram of sleep

Introduction The human electroencephalogram (EEG) of sleep undergoes profound changes with age, such as decreased sleep spindle amplitude and density in non-rapid eye movement stage 2 (NREM2). However, it is unknown how accurately a patient’s age can be predicted from EEG activity during sleep. A quantitative characterization of age-related EEG provides important insights into healthy aging. Moreover, the ability to detect deviations of EEG patterns from those typical for age could provide insights into age-related neurological disorders, and might provide a way to gauge the effects of interventions designed to preserve or improve brain health. Here we develop a model to predict a patient’s age based on large-scale and heterogeneous sleep EEG datasets. The prediction is called “brain age” (BA). Methods Datasets: (1) MGH sleep dataset: 3100 patients aged 18–80 years. (2) sleep-heart health study (SHHS): 3680 paired recordings aged 18–80 years, where each pair is recorded approximately 5 years apart from the same subject. This dataset is used to verify the longitudinal reliability of the brain age. EEG features: 102 features are extracted from 30s-epochs, and then averaged separately for the 5 sleep stages, yielding 510 features to summarize each patient’s overnight sleep. We also analyze medications and clinical variables to identify factors that help account for brain age being older or younger than chronological age. Results For 1000 testing patients from MGH dataset, the Pearson’s correlation between EEG-based brain age and chronological age is 0.86 (95% CI 0.84–0.87). The mean absolute prediction error (MAE) is 6.6 years. For SHHS dataset, training the model on a subset of 2000 records and testing on the other 1680 records achieves correlation at 0.71 (95% CI 0.69–0.73) and MAE 5.9 yrs. The average difference of BA between each pair is 4.8yrs. Training the entire MGH dataset and testing on SHHS achieves correlation at 0.61 (95% CI 0.59–0.63) and MAE 8.6 yrs. The average difference of BA between each pair is 3.7 yrs. In the MGH dataset, older brain age (predicted age greater than chronological age) is associated diabetes (Kruskal-Wallis test p-value 0.01) and weakly associated with wake time after sleep onset (Pearson’s correlation p-value 0.07). Conclusion Our results indicate that, at the population level, chronological age can be accurately predicted from overnight sleep EEG. Moreover, brain age accurately tracks chronological age. Further research is needed to characterize how EEG-based brain age relates to cognitive function and to what degree and by what means brain age is modifiable.