Predicting Brain Age Using Structural Neuroimaging and Deep Learning

Early detection of age-related diseases will greatly benefit from a model of the underlying biological aging process. In this paper, we develop a brain-age predictor by using structural magnetic resonance imaging (SMRI) and deep learning and evaluate the predicted brain age as a marker of brain-aging in Alzheimer's disease. Our approach does not require any domain knowledge in that it trains end-to-end on the SMRI image itself, and has been validated on real SMRI data collected from elderly subjects. We developed two different models by using convolutional neural network (CNN) based regression and bucket classification to predict brain ages from SMRI images. Our models achieved root mean squared errors (RMSE) of 5.54 and 6.44 years in predicting brain ages of healthy subjects. Further analysis showed that there is a substantial difference between the predicted brain ages of cognitively impaired and healthy subjects with similar chronological ages.

Download Full-text

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

10.1101/712851 ◽

2019 ◽

Cited By ~ 2

Author(s):

Maxwell L. Elliott ◽

Daniel W. Belsky ◽

Annchen R. Knodt ◽

David Ireland ◽

Tracy R. Melzer ◽

...

Keyword(s):

Cognitive Decline ◽

Birth Cohort ◽

Brain Aging ◽

Chronological Age ◽

Biological Aging ◽

Imaging Data ◽

Cohort Differences ◽

Age Related ◽

Brain Age ◽

The Brain

AbstractAn individual’s brain-age is the difference between chronological age and age predicted from machine-learning models of brain-imaging data. Brain-age has been proposed as a biomarker of age-related deterioration of the brain. Having an older brain-age 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 brain-age a biomarker of accelerated brain aging, a study is needed of a large cohort all born the same year who nevertheless vary on brain-age. In a population-representative 1972-73 birth cohort we measured brain-age at age 45, 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, brain-age was measured reliably (ICC=.81) and ranged from 24 to 72 years. Those with older midlife brain-ages tended to have poorer cognitive function in both adulthood and childhood, as well as impaired brain health at age 3. Furthermore, those with older brain-ages 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 brain-age as a potential surrogate biomarker for midlife intervention studies that seek to measure treatment response to dementia-prevention efforts in midlife. However, the findings also caution against the assumption that brain-age scores represent only age-related deterioration of the brain as they may also index central nervous system variation present since childhood.

Download Full-text

Exploring Epigenetic Age in Response to Intensive Relaxing Training: A Pilot Study to Slow Down Biological Age

International Journal of Environmental Research and Public Health ◽

10.3390/ijerph16173074 ◽

2019 ◽

Vol 16 (17) ◽

pp. 3074 ◽

Cited By ~ 5

Author(s):

Pavanello ◽

Campisi ◽

Tona ◽

Lin ◽

Iliceto

Keyword(s):

Myocardial Infarction ◽

Age Estimation ◽

Healthy Subjects ◽

Molecular Mechanisms ◽

Biological Age ◽

Chronological Age ◽

Biological Aging ◽

Epigenetic Clock ◽

Age Related ◽

Epigenetic Age

DNA methylation (DNAm) is an emerging estimator of biological aging, i.e., the often-defined “epigenetic clock”, with a unique accuracy for chronological age estimation (DNAmAge). In this pilot longitudinal study, we examine the hypothesis that intensive relaxing training of 60 days in patients after myocardial infarction and in healthy subjects may influence leucocyte DNAmAge by turning back the epigenetic clock. Moreover, we compare DNAmAge with another mechanism of biological age, leucocyte telomere length (LTL) and telomerase. DNAmAge is reduced after training in healthy subjects (p = 0.053), but not in patients. LTL is preserved after intervention in healthy subjects, while it continues to decrease in patients (p = 0.051). The conventional negative correlation between LTL and chronological age becomes positive after training in both patients (p < 0.01) and healthy subjects (p < 0.05). In our subjects, DNAmAge is not associated with LTL. Our findings would suggest that intensive relaxing practices influence different aging molecular mechanisms, i.e., DNAmAge and LTL, with a rejuvenating effect. Our study reveals that DNAmAge may represent an accurate tool to measure the effectiveness of lifestyle-based interventions in the prevention of age-related diseases.

Download Full-text

Influence of APOE genotype in primary age-related tauopathy

Acta Neuropathologica Communications ◽

10.1186/s40478-020-01095-1 ◽

2020 ◽

Vol 8 (1) ◽

Author(s):

Andrew C. Robinson ◽

Yvonne S. Davidson ◽

Federico Roncaroli ◽

James Minshull ◽

Phillip Tinkler ◽

...

Keyword(s):

Apoe Genotype ◽

Protein Isoforms ◽

Elderly Subjects ◽

Cognitively Impaired ◽

Age Related ◽

Dementia Research ◽

Amyloid Aβ ◽

The University ◽

University Of Manchester ◽

The Impact

AbstractThe term “Primary age-related tauopathy” (PART) was coined in 2014 to describe the common neuropathological observation of neurofibrillary tangles without associated beta-amyloid (Aβ) pathology. It is possible for PART pathology to be present in both cognitively normal and cognitively impaired individuals. Genetically, Apolipoprotein E (APOE) ε4 has been shown to occur less commonly in PART than in Alzheimer’s disease (AD). Here, we investigate the relationships between PART, AD and those pathologically normal for age, with an emphasis on APOE and cognition, using 152 selected participants from The University of Manchester Longitudinal Study of Cognition in Normal Healthy Old Age and the Manchester arm of the Brains for Dementia Research cohort. APOE genotype differed between PART and AD with APOE ε2 more common in the former and APOE ε4 more common in the latter. Individuals with definite PART were less likely to be cognitively impaired than those with AD and those with pathology considered pathologically normal for age. We postulate that the lack of Aβ in definite PART cases may be due either to an increased frequency of APOE ε2 or decreased frequency of APOE ε4 as their resulting protein isoforms have differing binding properties in relation to Aβ. Similarly, an increased frequency of APOE ε2 or decreased frequency of APOE ε4 may lead to decreased levels of cognitive impairment, which raises questions regarding the impact of Aβ pathology on overall cognition in elderly subjects. We suggest that it may be possible to use the increased frequency of APOE ε2 in definite PART to assist neuropathological diagnosis.

Download Full-text

Dietary habits are major determinants of the plasma antioxidant status in healthy elderly subjects

British Journal Of Nutrition ◽

10.1079/bjn20051574 ◽

2005 ◽

Vol 94 (5) ◽

pp. 639-642 ◽

Cited By ~ 42

Author(s):

Timur Anlasik ◽

Helmut Sies ◽

Helen R. Griffiths ◽

Patrizia Mecocci ◽

Wilhelm Stahl ◽

...

Keyword(s):

Plasma Levels ◽

Healthy Subjects ◽

Antioxidant Status ◽

Fruits And Vegetables ◽

Community Dwelling ◽

Protein Carbonyls ◽

Elderly Subjects ◽

Age And Gender ◽

Age Related ◽

And Gender

Previous studies indicate that regular consumption of a diet rich in fruits and vegetables is associated with a lower risk for age-related diseases. The aim of the present study was to evaluate whether the often-reported age-related decrease of plasma antioxidants in man depends on differences in dietary intake or on other age- and gender-related factors. In this observational case-control study, thirty-nine community-dwelling healthy subjects aged 65 years and older consuming high intakes of fruits and vegetables daily (HI) and forty-eight healthy subjects aged 65 and older consuming low intakes of fruit and vegetables daily (LI) were enrolled. Plasma levels of retinol, tocopherols, carotenoids and malondialdehyde (MDA) as well as content of protein carbonyls in Ig G were measured. Plasma levels of retinol, tocopherols and carotenoids were significantly higher in group HI than in group LI subjects independent of age and gender. MDA levels were inversely correlated with vitamin A and α-carotene. Protein carbonyls were inversely correlated with γ-tocopherol. In the elderly, a higher daily intake of fruits and vegetables is associated with an improved antioxidant status in comparison to subjects consuming diets poor in fruits and vegetables. Modification of nutritional habits among other lifestyle changes should be encouraged to lower prevalence of disease risk factors in later life.

Download Full-text

Age-Related Changes in Distribution of Body Weight on Soles of Feet for Selected Actions and Postures

Perceptual and Motor Skills ◽

10.2466/pms.1997.85.3f.1263 ◽

1997 ◽

Vol 85 (3_suppl) ◽

pp. 1263-1271 ◽

Cited By ~ 1

Author(s):

Toshiaki Yanagida ◽

Takaaki Asami

Keyword(s):

Body Weight ◽

Weight Distribution ◽

Healthy Subjects ◽

Age Groups ◽

The Body ◽

The Other ◽

Elderly Subjects ◽

Age Related ◽

Soles Of The Feet ◽

Age Related Changes

We investigated age-related changes in the distribution of body weight on soles of feet in 878 healthy subjects ranging from 5 to 80 years of age. By modifying Morton's Staticometer, we constructed an instrument for measuring body-weight distribution over three areas of soles of the feet, the big toe (inner forefoot), the other four toes combined (outer forefoot) and the heels for both feet, thus a total of six areas. The weights in the six areas were recorded at the completion of nine selected actions and postures. We observed that for inhaling and exhaling standing postures, generally younger subjects had a ratio close to 1:2:3 for weights recorded for the inner toe:outer toes:heels as observed by Morton, but elderly subjects had a smaller value than 3 for the heel. The body-weight distribution tended to shift from heels to outer toes across age groups, which was more distinctly observed in women than in men.

Download Full-text

Deep learning based quantification of the accelerated brain aging rate in glioma patients after radiotherapy

10.1101/2021.12.22.21267899 ◽

2021 ◽

Author(s):

Selena I. Huisman ◽

Arthur T.J. van der Boog ◽

Fia Cialdella ◽

Joost J.C. Verhoeff ◽

Szabolcs David

Keyword(s):

Deep Learning ◽

Brain Aging ◽

Accelerated Aging ◽

Learning Model ◽

Aging Rate ◽

Saliency Maps ◽

Mri Scans ◽

Brain Age ◽

Deep Learning Model

Background and purpose. Changes of healthy appearing brain tissue after radiotherapy have been previously observed, however, they remain difficult to quantify. Due to these changes, patients undergoing radiotherapy may have a higher risk of cognitive decline, leading to a reduced quality of life. The experienced tissue atrophy is similar to the effects of normal aging in healthy individuals. We propose a new way to quantify tissue changes after cranial RT as accelerated brain aging using the BrainAGE framework. Materials and methods. BrainAGE was applied to longitudinal MRI scans of 32 glioma patients, who have undergone radiotherapy. Utilizing a pre-trained deep learning model, brain age is estimated for all patients' pre-radiotherapy planning and follow-up MRI scans to get a quantification of the changes occurring in the brain over time. Saliency maps were extracted from the model to spatially identify which areas of the brain the deep learning model weighs highest for predicting age. The predicted ages from the deep learning model were used in a linear mixed effects model to quantity aging and aging rates for patients after radiotherapy. Results. The linear mixed effects model resulted in an accelerated aging rate of 2.78 years per year, a significant increase over a normal aging rate of 1 (p <0.05, confidence interval (CI) = 2.54-3.02). Furthermore, the saliency maps showed numerous anatomically well-defined areas, e.g.: Heschl's gyrus among others, determined by the model as important for brain age prediction. Conclusion. We found that patients undergoing radiotherapy are affected by significant radiation- induced accelerated aging, with several anatomically well-defined areas contributing to this aging. The estimated brain age could provide a method for quantifying quality of life post-radiotherapy.

Download Full-text

Using deep learning to predict brain age from brain magnetic resonance images and cognitive tests reveals that anatomical and functional brain aging are phenotypically and genetically distinct

10.1101/2021.06.22.21259280 ◽

2021 ◽

Author(s):

Alan Le Goallec ◽

Samuel Diai ◽

Sasha Collin ◽

Theo Vincent ◽

Chirag J Patel

Keyword(s):

Magnetic Resonance ◽

Genetic Factors ◽

Brain Aging ◽

Magnetic Resonance Images ◽

Brain Diseases ◽

World Population ◽

Cognitive Tests ◽

Functional Brain ◽

Age Related ◽

Brain Age

With the world population aging, the prevalence of age-related brain diseases such as Alzheimer's, Parkinson's, Lou Gehrig's, and cerebrovascular diseases. In the following, we built brain age predictors by leveraging 46,000 brain magnetic resonance images [MRIs] and cognitive tests from UK Biobank participants. We predicted age with a R-Squared [R2] of 76.4+/-1.0% and a root mean squared error of 3.58+/-0.05 years. We defined accelerated brain aging as the difference between brain age (predicted age) and age. Accelerated brain aging is partially heritable (h_g2=35.9+/-2.6%), and is associated with 219 single nucleotide polymorphisms [SNPs] in 25 genes (e.g CRHR1, involved in the hypothalamic-pituitary-adrenal pathway). Similarly, it is associated with biomarkers (e.g blood pressure), clinical phenotypes (e.g general health), diseases (e.g diabetes), environmental (e.g smoking) and socioeconomic variables (e.g income and education). We performed the same analysis, this time distinguishing between anatomical (MRI-based) and functional (cognitive tests-based) brain aging. We found the two accelerated aging phenotypes to be phenotypically .112+/-.006 correlated and genetically uncorrelated, with distinct SNPs and non-genetic factors associated with each. In conclusion, anatomical and functional brain aging are two distinct, complex phenotypes, which also differ in their genetic and non-genetic factors. Our brain predictors could be used to monitor the effects of emerging rejuvenating therapies on the brain.

Download Full-text

Longitudinal trajectories of brain age in young individuals at familial risk of mood disorder from the Scottish Bipolar Family Study

Wellcome Open Research ◽

10.12688/wellcomeopenres.15617.3 ◽

2020 ◽

Vol 4 ◽

pp. 206

Author(s):

Laura de Nooij ◽

Mathew A. Harris ◽

Emma L. Hawkins ◽

Toni-Kim Clarke ◽

Xueyi Shen ◽

...

Keyword(s):

Mood Disorder ◽

Brain Structure ◽

Family Study ◽

Brain Aging ◽

Familial Risk ◽

Age Difference ◽

Age Related ◽

Prospective Longitudinal ◽

Brain Age

Background: Within young individuals, mood disorder onset may be related to changes in trajectory of brain structure development. To date, however, longitudinal prospective studies remain scarce and show partly contradictory findings, with a lack of emphasis on changes at the level of global brain patterns. Cross-sectional adult studies have applied such methods and show that mood disorders are associated with accelerated brain aging. Currently, it remains unclear whether young individuals show differential brain structure aging trajectories associated with onset of mood disorder and/or presence of familial risk. Methods: Participants included young individuals (15-30 years, 53%F) from the prospective longitudinal Scottish Bipolar Family Study with and without close family history of mood disorder. All were well at time of recruitment. Implementing a structural MRI-based brain age prediction model, we globally assessed individual trajectories of age-related structural change using the difference between predicted brain age and chronological age (brain-predicted age difference (brain-PAD)) at baseline and at 2-year follow-up. Based on follow-up clinical assessment, individuals were categorised into three groups: (i) controls who remained well (C-well, n = 93), (ii) high familial risk who remained well (HR-well, n = 74) and (iii) high familial risk who developed a mood disorder (HR-MD, n = 35). Results: At baseline, brain-PAD was comparable between groups. Results showed statistically significant negative trajectories of brain-PAD between baseline and follow-up for HR-MD versus C-well (β = -0.60, pcorrected < 0.001) and HR-well (β = -0.36, pcorrected = 0.02), with a potential intermediate trajectory for HR-well (β = -0.24 years, pcorrected = 0.06). Conclusions: These preliminary findings suggest that within young individuals, onset of mood disorder and familial risk may be associated with a deceleration in brain structure aging trajectories. Extended longitudinal research will need to corroborate findings of emerging maturational lags in relation to mood disorder risk and onset.

Download Full-text

Deep learning-based brain age prediction in normal aging and dementia

10.21203/rs.3.rs-804454/v1 ◽

2021 ◽

Author(s):

Jeyeon Lee ◽

Brian Burkett ◽

Hoon-Ki Min ◽

Matthew Senjem ◽

Emily Lundt ◽

...

Keyword(s):

Deep Learning ◽

Cognitive Impairment ◽

Brain Aging ◽

Lewy Body Dementia ◽

Normal Aging ◽

Specific Pattern ◽

Normal Brain ◽

Age Gap ◽

Brain Age ◽

Age Prediction

Abstract Normal brain aging is accompanied by patterns of functional and structural change. Alzheimer's disease (AD), a representative neurodegenerative disease, has been linked to accelerated brain aging at respective age ranges. Here, we developed a deep learning-based brain age prediction model using fluorodeoxyglucose (FDG) PET and structural MRI and tested how the brain age gap relates to degenerative cognitive syndromes including mild cognitive impairment, AD, frontotemporal dementia, and Lewy body dementia. Occlusion analysis, performed to facilitate interpretation of the model, revealed that the model learns an age- and modality-specific pattern of brain aging. The elevated brain age gap in dementia cohorts was highly correlated with the cognitive impairment and AD biomarker. However, regions generating brain age gaps were different for each diagnosis group of which the AD continuum showed similar patterns to normal aging in the CU.

Download Full-text

Quantification of biological aging in young adults

Proceedings of the National Academy of Sciences ◽

10.1073/pnas.1506264112 ◽

2015 ◽

Vol 112 (30) ◽

pp. E4104-E4110 ◽

Cited By ~ 306

Author(s):

Daniel W. Belsky ◽

Avshalom Caspi ◽

Renate Houts ◽

Harvey J. Cohen ◽

David L. Corcoran ◽

...

Keyword(s):

Young Adults ◽

Brain Aging ◽

Multiple Organ ◽

Biological Aging ◽

Research Translation ◽

Human Aging ◽

Aging Research ◽

Cross Sectional ◽

Age Related ◽

Organ Systems

Antiaging therapies show promise in model organism research. Translation to humans is needed to address the challenges of an aging global population. Interventions to slow human aging will need to be applied to still-young individuals. However, most human aging research examines older adults, many with chronic disease. As a result, little is known about aging in young humans. We studied aging in 954 young humans, the Dunedin Study birth cohort, tracking multiple biomarkers across three time points spanning their third and fourth decades of life. We developed and validated two methods by which aging can be measured in young adults, one cross-sectional and one longitudinal. Our longitudinal measure allows quantification of the pace of coordinated physiological deterioration across multiple organ systems (e.g., pulmonary, periodontal, cardiovascular, renal, hepatic, and immune function). We applied these methods to assess biological aging in young humans who had not yet developed age-related diseases. Young individuals of the same chronological age varied in their “biological aging” (declining integrity of multiple organ systems). Already, before midlife, individuals who were aging more rapidly were less physically able, showed cognitive decline and brain aging, self-reported worse health, and looked older. Measured biological aging in young adults can be used to identify causes of aging and evaluate rejuvenation therapies.

Download Full-text