Brain Age Estimation from Overnight Sleep Electroencephalography with Multi-Flow Sequence Learning.

Purpose: This study aims to improve brain age estimation by developing a novel deep learning model utilizing overnight electroencephalography (EEG) data. Methods: We address limitations in current brain age prediction methods by proposing a model trained and evaluated on multiple cohort data, covering a broad age range. The model employs a one-dimensional Swin Transformer to efficiently extract complex patterns from sleep EEG signals and a convolutional neural network with attentional mechanisms to summarize sleep structural features. A multi-flow learning-based framework attentively merges these two features, employing sleep structural information to direct and augment the EEG features. A post-prediction model is designed to integrate the age-related features throughout the night. Furthermore, we propose a DecadeCE loss function to address the problem of an uneven age distribution. Results: We utilized 18,767 polysomnograms (PSGs) from 13,616 subjects to develop and evaluate the proposed model. The model achieves a mean absolute error (MAE) of 4.19 and a correlation of 0.97 on the mixed-cohort test set, and an MAE of 6.18 years and a correlation of 0.78 on an independent test set. Our brain age estimation work reduced the error by more than 1 year compared to other studies that also used EEG, achieving the level of neuroimaging. The estimated brain age index demonstrated longitudinal sensitivity and exhibited a significant increase of 1.27 years in individuals with psychiatric or neurological disorders relative to healthy individuals. Conclusion: The multi-flow deep learning model proposed in this study, based on overnight EEG, represents a more accurate approach for estimating brain age. The utilization of overnight sleep EEG for the prediction of brain age is both cost-effective and adept at capturing dynamic changes. These findings demonstrate the potential of EEG in predicting brain age, presenting a noninvasive and accessible method for assessing brain aging.

Predictive values of pre-treatment brain age models to rTMS effects in neurocognitive disorder with depression: Secondary analysis of a randomised sham-controlled clinical trial.

INTRODUCTION: One major challenge in developing personalised repetitive transcranial magnetic stimulation (rTMS) is that the treatment responses exhibited high inter-individual variations. Brain morphometry might contribute to these variations. This study sought to determine whether individual's brain morphometry could predict the rTMS responders and remitters. METHODS: This was a secondary analysis of data from a randomised clinical trial that included fifty-five patients over the age of 60 with both comorbid depression and neurocognitive disorder. Based on magnetic resonance imaging scans, estimated brain age was calculated with morphometric features using a support vector machine. Brain-predicted age difference (brain-PAD) was computed as the difference between brain age and chronological age. RESULTS: The rTMS responders and remitters had younger brain age. Every additional year of brain-PAD decreased the odds of relieving depressive symptoms by ∼25.7% in responders (Odd ratio [OR] = 0.743, p = .045) and by ∼39.5% in remitters (OR = 0.605, p = .022) in active rTMS group. Using brain-PAD score as a feature, responder-nonresponder classification accuracies of 85% (3rd week) and 84% (12th week), respectively were achieved. CONCLUSION: In elderly patients, younger brain age appears to be associated with better treatment responses to active rTMS. Pre-treatment brain age models informed by morphometry might be used as an indicator to stratify suitable patients for rTMS treatment. TRIAL REGISTRATION: ClinicalTrials.gov Identifier: ChiCTR-IOR-16008191.

The role of cortical structural variance in deep learning-based prediction of fetal brain age.

Background: Deep-learning-based brain age estimation using magnetic resonance imaging data has been proposed to identify abnormalities in brain development and the risk of adverse developmental outcomes in the fetal brain. Although saliency and attention activation maps have been used to understand the contribution of different brain regions in determining brain age, there has been no attempt to explain the influence of shape-related cortical structural features on the variance of predicted fetal brain age. Methods: We examined the association between the predicted brain age difference (PAD: predicted brain age-chronological age) from our convolution neural networks-based model and global and regional cortical structural measures, such as cortical volume, surface area, curvature, gyrification index, and folding depth, using regression analysis. Results: Our results showed that global brain volume and surface area were positively correlated with PAD. Additionally, higher cortical surface curvature and folding depth led to a significant increase in PAD in specific regions, including the perisylvian areas, where dramatic agerelated changes in folding structures were observed in the late second trimester. Furthermore, PAD decreased with disorganized sulcal area patterns, suggesting that the interrelated arrangement and areal patterning of the sulcal folds also significantly affected the prediction of fetal brain age. Conclusion: These results allow us to better understand the variance in deep learning-based fetal brain age and provide insight into the mechanism of the fetal brain age prediction model.

Brain age has limited utility as a biomarker for capturing fluid cognition in older individuals.

One well-known biomarker candidate that supposedly helps capture fluid cognition is Brain Age, or a predicted value based on machine-learning models built to predict chronological age from brain MRI. To formally evaluate the utility of Brain Age for capturing fluid cognition, we built 26 age-prediction models for Brain Age based on different combinations of MRI modalities, using the Human Connectome Project in Aging (n=504, 36-100 years old). First, based on commonality analyses, we found a large overlap between Brain Age and chronological age: Brain Age could uniquely add only around 1.6% in explaining variation in fluid cognition over and above chronological age. Second, the age-prediction models that performed better at predicting chronological age did NOT necessarily create better Brain Age for capturing fluid cognition over and above chronological age. Instead, better-performing age-prediction models created Brain Age that overlapped larger with chronological age, up to around 29% out of 32%, in explaining fluid cognition. Third, Brain Age missed around 11% of the total variation in fluid cognition that could have been explained by the brain variation. That is, directly predicting fluid cognition from brain MRI data (instead of relying on Brain Age and chronological age) could lead to around a 1/3-time improvement of the total variation explained. Accordingly, we demonstrated the limited utility of Brain Age as a biomarker for fluid cognition and made some suggestions to ensure the utility of Brain Age in explaining fluid cognition and other phenotypes of interest.

Brain reserve in midlife is associated with executive function changes across 12 years.

We examined how brain reserve in midlife, measured by brain-predicted age difference scores (Brain-PADs), predicted executive function concurrently and longitudinally into early old age, and whether these associations were moderated by young adult cognitive reserve or APOE genotype. 508 men in the Vietnam Era Twin Study of Aging (VETSA) completed neuroimaging assessments at mean age 56 and six executive function tasks at mean ages 56, 62, and 68 years. Results indicated that greater brain reserve at age 56 was associated with better concurrent executive function (r=.10, p=.040) and less decline in executive function over 12 years (r=.34, p=.001). These associations were not moderated by cognitive reserve or APOE genotype. Twin analysis suggested associations with executive function slopes were driven by genetic influences. Our findings suggest that greater brain reserve allowed for better cognitive maintenance from middle- to old age, driven by a genetic association. The results are consistent with differential preservation of executive function based on brain reserve that is independent of young adult cognitive reserve or APOE genotype.

Greater physical fitness ( VO 2 max $$ {\mathrm{VO}}_{2_{\mathrm{max}}} $$ ) in healthy older adults associated with increased integrity of the locus coeruleus-noradrenergic system.

AIM: Physical activity (PA) is a key component for brain health and Reserve, and it is among the main dementia protective factors. However, the neurobiological mechanisms underpinning Reserve are not fully understood. In this regard, a noradrenergic (NA) theory of cognitive reserve (Robertson, 2013) has proposed that the upregulation of NA system might be a key factor for building reserve and resilience to neurodegeneration because of the neuroprotective role of NA across the brain. PA elicits an enhanced catecholamine response, in particular for NA. By increasing physical commitment, a greater amount of NA is synthetised in response to higher oxygen demand. More physically trained individuals show greater capabilities to carry oxygen resulting in greater Vo 2 max $$ {\mathrm{Vo}}_{2_{\mathrm{max}}} $$ - a measure of oxygen uptake and physical fitness (PF). METHODS: We hypothesized that greater Vo 2 max $$ {\mathrm{Vo}}_{2_{\mathrm{max}}} $$ would be related to greater Locus Coeruleus (LC) MRI signal intensity. In a sample of 41 healthy subjects, we performed Voxel-Based Morphometry analyses, then repeated for the other neuromodulators as a control procedure (Serotonin, Dopamine and Acetylcholine). RESULTS: As hypothesized, greater Vo 2 max $$ {\mathrm{Vo}}_{2_{\mathrm{max}}} $$ related to greater LC signal intensity, and weaker associations emerged for the other neuromodulators. CONCLUSION: This newly established link between Vo 2 max $$ {\mathrm{Vo}}_{2_{\mathrm{max}}} $$ and LC-NA system offers further understanding of the neurobiology underpinning Reserve in relationship to PA. While this study supports Robertson's theory proposing the upregulation of the NA system as a possible key factor building Reserve, it also provides ground for increasing LC-NA system resilience to neurodegeneration via Vo 2 max $$ {\mathrm{Vo}}_{2_{\mathrm{max}}} $$ enhancement.

Predicted Brain Age in First-Episode Psychosis: Association with Inexpressivity.

Accelerated brain aging is a possible mechanism of pathology in schizophrenia. Advances in MRI-based brain development algorithms allow for the calculation of predicted brain age (PBA) for individuals. Here, we assessed PBA in 70 first-episode schizophrenia-spectrum individuals (FESz) and 76 matched healthy neurotypical comparison individuals (HC) to determine if FESz showed advanced aging proximal to psychosis onset and whether PBA was associated with neurocognitive, social functioning, or symptom severity measures. PBA was calculated with BrainAgeR (v2.1) from T1-weighted MR scans. There were no differences in the PBAs between groups. After controlling for actual age, a "younger" PBA was associated with higher vocabulary scores among all individuals, while an "older" PBA was associated with more severe negative symptom "Inexpressivity" component scores among FESz. Female participants in both groups had an elevated PBA relative to male participants. These results suggest that a relatively younger brain age is associated with a better semantic memory performance. There is no evidence for accelerated aging in FESz with a late adolescent/early adult onset. Despite a normative PBA, FESz with a greater residual PBA showed impairments in a cluster of negative symptoms, which may indicate some underlying age-related pathology proximal to psychosis onset. Although a period of accelerated aging cannot be ruled out with disease course, it does not occur at the time of the first episode.

Tau Protein Accumulation Trajectory-Based Brain Age Prediction in the Alzheimer's Disease Continuum.

Clinical cognitive advancement within the Alzheimer's disease (AD) continuum is intimately connected with sustained accumulation of tau protein pathology. The biological brain age and its gap show great potential for pathological risk and disease severity. In the present study, we applied multivariable linear support vector regression to train a normative brain age prediction model using tau brain images. We further assessed the predicted biological brain age and its gap for patients within the AD continuum. In the AD continuum, evaluated pathologic tau binding was found in the inferior temporal, parietal-temporal junction, precuneus/posterior cingulate, dorsal frontal, occipital, and inferior-medial temporal cortices. The biological brain age gaps of patients within the AD continuum were notably higher than those of the normal controls (p < 0.0001). Significant positive correlations were observed between the brain age gap and global tau protein accumulation levels for mild cognitive impairment (r = 0.726, p < 0.001), AD (r = 0.845, p < 0.001), and AD continuum (r = 0.797, p < 0.001). The pathologic tau-based age gap was significantly linked to neuropsychological scores. The proposed pathologic tau-based biological brain age model could track the tau protein accumulation trajectory of cognitive impairment and further provide a comprehensive quantification index for the tau accumulation risk.

Functional network centrality indicates interactions between APOE4 and age across the clinical spectrum of AD.

Advanced age is the most important risk factor for Alzheimer's disease (AD), and carrier-status of the Apolipoprotein E4 (APOE4) allele is the strongest known genetic risk factor. Many studies have consistently shown a link between APOE4 and synaptic dysfunction, possibly reflecting pathologically accelerated biological aging in persons at risk for AD. To test the hypothesis that distinct functional connectivity patterns characterize APOE4 carriers across the clinical spectrum of AD, we investigated 128 resting state functional Magnetic Resonance Imaging (fMRI) datasets from the Alzheimer's Disease Neuroimaging Initiative database (ADNI), representing all disease stages from cognitive normal to clinical dementia. Brain region centralities within functional networks, computed as eigenvector centrality, were tested for multivariate associations with chronological age, APOE4 carrier status and clinical stage (as well as their interactions) by partial least square analysis (PLSC). By PLSC analysis two distinct brain activity patterns could be identified, which reflected interactive effects of age, APOE4 and clinical disease stage. A first component including sensorimotor regions and parietal regions correlated with age and AD clinical stage (p < 0.001). A second component focused on medial-frontal regions and was specifically related to the interaction between age and APOE4 (p = 0.032). Our findings are consistent with earlier reports on altered network connectivity in APOE4 carriers. Results of our study highlight promise of graph-theory based network centrality to identify brain connectivity linked to genetic risk, clinical stage and age. Our data suggest the existence of brain network activity patterns that characterize APOE4 carriers across clinical stages of AD.

Linking menopause-related factors, history of depression, APOE ε4, and proxies of biological aging in the UK biobank cohort.

In a subset of females, postmenopausal status has been linked to accelerated aging and neurological decline. A complex interplay between reproductive-related factors, mental disorders, and genetics may influence brain function and accelerate the rate of aging in the postmenopausal phase. Using multiple regressions corrected for age, in this preregistered study we investigated the associations between menopause-related factors (i.e., menopausal status, menopause type, age at menopause, and reproductive span) and proxies of cellular aging (leukocyte telomere length, LTL) and brain aging (white and gray matter brain age gap, BAG) in 13,780 females from the UK Biobank (age range 39-82). We then determined how these proxies of aging were associated with each other, and evaluated the effects of menopause-related factors, history of depression (= lifetime broad depression), and APOE ε4 genotype on BAG and LTL, examining both additive and interactive relationships. We found that postmenopausal status and older age at natural menopause were linked to longer LTL and lower BAG. Surgical menopause and longer natural reproductive span were also associated with longer LTL. BAG and LTL were not significantly associated with each other. The greatest variance in each proxy of biological aging was most consistently explained by models with the addition of both lifetime broad depression and APOE ε4 genotype. Overall, this study demonstrates a complex interplay between menopause-related factors, lifetime broad depression, APOE ε4 genotype, and proxies of biological aging. However, results are potentially influenced by a disproportionate number of healthier participants among postmenopausal females. Future longitudinal studies incorporating heterogeneous samples are an essential step towards advancing female health.

Prediction of chronological age from resting-state EEG power in the first three years of life.

The infant brain undergoes rapid and significant developmental changes in the first three years of life. Understanding these changes through the prediction of chronological age using neuroimaging data can provide insights into typical and atypical brain development. We utilized longitudinal resting-state EEG data from 457 typically developing infants, comprising 938 recordings, to develop age prediction models. The multilayer perceptron model demonstrated the highest accuracy with an R2 of 0.82 and a mean absolute error of 92.4 days. Aperiodic offset and periodic theta, alpha, and beta power were identified as key predictors of age via Shapley values. Application of the model to EEG data from infants later diagnosed with autism spectrum disorder or Down syndrome revealed significant underestimations of chronological age. This study establishes the feasibility of using EEG to assess brain maturation in early childhood and supports its potential as a clinical tool for early identification of alterations in brain development.

[18F]FDG PET integrated with structural MRI for accurate brain age prediction.

PURPOSE: Brain aging is a complex and heterogeneous process characterized by both structural and functional decline. This study aimed to establish a novel deep learning (DL) method for predicting brain age by utilizing structural and metabolic imaging data. METHODS: The dataset comprised participants from both the Universal Medical Imaging Diagnostic Center (UMIDC) and the Alzheimer's Disease Neuroimaging Initiative (ADNI). The former recruited 395 normal control (NC) subjects, while the latter included 438 NC subjects, 51 mild cognitive impairment (MCI) subjects, and 56 Alzheimer's disease (AD) subjects. We developed a novel dual-pathway, 3D simple fully convolutional network (Dual-SFCNeXt) to estimate brain age using [18F]fluorodeoxyglucose positron emission tomography ([18F]FDG PET) and structural magnetic resonance imaging (sMRI) images of NC subjects as input. Several prevailing DL models were trained and tested using either MRI or PET data for comparison. Model accuracies were evaluated using mean absolute error (MAE) and Pearson's correlation coefficient (r). Brain age gap (BAG), deviations of brain age from chronologic age, was correlated with cognitive assessments in MCI and AD subjects. RESULTS: Both PET- and MRI-based models achieved high prediction accuracy. The leading model was the SFCNeXt (the single-pathway version) for PET (MAE = 2.92, r = 0.96) and MRI (MAE = 3.23, r = 0.95) on all samples. By integrating both PET and MRI images, the Dual-SFCNeXt demonstrated significantly improved accuracy (MAE = 2.37, r = 0.97) compared to all single-modality models. Significantly higher BAG was observed in both the AD (P < 0.0001) and MCI (P < 0.0001) groups compared to the NC group. BAG correlated significantly with Mini-Mental State Examination (MMSE) scores (r=-0.390 for AD, r=-0.436 for MCI) and the Clinical Dementia Rating Scale Sum of Boxes (CDR-SB) scores (r = 0.333 for AD, r = 0.372 for MCI). CONCLUSION: The integration of [18F]FDG PET with structural MRI enhances the accuracy of brain age prediction, potentially introducing a new avenue for related multimodal brain age prediction studies.

Brain age gap estimation using attention-based ResNet method for Alzheimer's disease detection.

This study investigates the correlation between brain age and chronological age in healthy individuals using brain MRI images, aiming to identify potential biomarkers for neurodegenerative diseases like Alzheimer's. To achieve this, a novel attention-based ResNet method, 3D-Attention-Resent-SVR, is proposed to accurately estimate brain age and distinguish between Cognitively Normal (CN) and Alzheimer's disease (AD) individuals by computing the brain age gap (BAG). Unlike conventional methods, which often rely on single datasets, our approach addresses potential biases by employing four datasets for training and testing. The results, based on a combined dataset from four public sources comprising 3844 data points, demonstrate the model's efficacy with a mean absolute error (MAE) of 2.05 for brain age gap estimation. Moreover, the model's generalizability is showcased by training on three datasets and testing on a separate one, yielding a remarkable MAE of 2.4. Furthermore, leveraging BAG as the sole biomarker, our method achieves an accuracy of 92% and an AUC of 0.87 in Alzheimer's disease detection on the ADNI dataset. These findings underscore the potential of our approach in assisting with early detection and disease monitoring, emphasizing the strong correlation between BAG and AD.

Brain age as a biomarker for pathological versus healthy ageing - a REMEMBER study.

OBJECTIVES: This study aimed to evaluate the potential clinical value of a new brain age prediction model as a single interpretable variable representing the condition of our brain. Among many clinical use cases, brain age could be a novel outcome measure to assess the preventive effect of life-style interventions. METHODS: The REMEMBER study population (N = 742) consisted of cognitively healthy (HC,N = 91), subjective cognitive decline (SCD,N = 65), mild cognitive impairment (MCI,N = 319) and AD dementia (ADD,N = 267) subjects. Automated brain volumetry of global, cortical, and subcortical brain structures computed by the CE-labeled and FDA-cleared software icobrain dm (dementia) was retrospectively extracted from T1-weighted MRI sequences that were acquired during clinical routine at participating memory clinics from the Belgian Dementia Council. The volumetric features, along with sex, were combined into a weighted sum using a linear model, and were used to predict 'brain age' and 'brain predicted age difference' (BPAD = brain age-chronological age) for every subject. RESULTS: MCI and ADD patients showed an increased brain age compared to their chronological age. Overall, brain age outperformed BPAD and chronological age in terms of classification accuracy across the AD spectrum. There was a weak-to-moderate correlation between total MMSE score and both brain age (r = -0.38,p < .001) and BPAD (r = -0.26,p < .001). Noticeable trends, but no significant correlations, were found between BPAD and incidence of conversion from MCI to ADD, nor between BPAD and conversion time from MCI to ADD. BPAD was increased in heavy alcohol drinkers compared to non-/sporadic (p = .014) and moderate (p = .040) drinkers. CONCLUSIONS: Brain age and associated BPAD have the potential to serve as indicators for, and to evaluate the impact of lifestyle modifications or interventions on, brain health.

The impact of age-related hearing loss on cognitive decline: The mediating role of brain age gap.

In recent years, the relationship between age-related hearing loss, cognitive decline, and the risk of dementia has garnered significant attention. The significant variability in brain health and aging among individuals of the same chronological age suggests that a measure assessing how one's brain ages may better explain hearing-cognition links. The main aim of this study was to investigate the mediating role of Brain Age Gap (BAG) in the association between hearing impairment and cognitive function. This research included 185 participants aged 20-79 years. BAG was estimated based on the difference between participant's brain age (estimated based on their structural T1-weighted MRI scans) and chronological age. Cognitive performance was assessed using the Montreal Cognitive Assessment (MoCA) test while hearing ability was measured using pure-tone thresholds (PTT) and words-in-noise (WIN) perception. Mediation analyses were used to examine the mediating role of BAG in the relationship between age-related hearing loss as well as difficulties in WIN perception and cognition. Participants with poorer hearing sensitivity and WIN perception showed lower MoCA scores, but this was an indirect effect. Participants with poorer performance on PTT and WIN tests had larger BAG (accelerated brain aging), and this was associated with poorer performance on the MoCA test. Mediation analyses showed that BAG partially mediated the relationship between age-related hearing loss and cognitive decline. This study enhances our understanding of the interplay among hearing loss, cognition, and BAG, emphasizing the potential value of incorporating brain age assessments in clinical evaluations to gain insights beyond chronological age, thus advancing strategies for preserving cognitive health in aging populations.

Brain-age estimation with a low-cost EEG-headset: effectiveness and implications for large-scale screening and brain optimization.

Over time, pathological, genetic, environmental, and lifestyle factors can age the brain and diminish its functional capabilities. While these factors can lead to disorders that can be diagnosed and treated once they become symptomatic, often treatment is difficult or ineffective by the time significant overt symptoms appear. One approach to this problem is to develop a method for assessing general age-related brain health and function that can be implemented widely and inexpensively. To this end, we trained a machine-learning algorithm on resting-state EEG (RS-EEG) recordings obtained from healthy individuals as the core of a brain-age estimation technique that takes an individual's RS-EEG recorded with the low-cost, user-friendly EMOTIV EPOC X headset and returns that person's estimated brain age. We tested the current version of our machine-learning model against an independent test-set of healthy participants and obtained a correlation coefficient of 0.582 between the chronological and estimated brain ages (r = 0.963 after statistical bias-correction). The test-retest correlation was 0.750 (0.939 after bias-correction) over a period of 1 week. Given these strong results and the ease and low cost of implementation, this technique has the potential for widespread adoption in the clinic, workplace, and home as a method for assessing general brain health and function and for testing the impact of interventions over time.

Resting state electroencephalographic brain activity in neonates can predict age and is indicative of neurodevelopmental outcome.

OBJECTIVE: Electroencephalography (EEG) can be used to estimate neonates' biological brain age. Discrepancies between postmenstrual age and brain age, termed the brain age gap, can potentially quantify maturational deviation. Existing brain age EEG models are not well suited to clinical cot-side use for estimating neonates' brain age gap due to their dependency on relatively large data and pre-processing requirements. METHODS: We trained a deep learning model on resting state EEG data from preterm neonates with normal neurodevelopmental Bayley Scale of Infant and Toddler Development (BSID) outcomes, using substantially reduced data requirements. We subsequently tested this model in two independent datasets from two clinical sites. RESULTS: In both test datasets, using only 20 min of resting-state EEG activity from a single channel, the model generated accurate age predictions: mean absolute error = 1.03 weeks (p-value = 0.0001) and 0.98 weeks (p-value = 0.0001). In one test dataset, where 9-month follow-up BSID outcomes were available, the average neonatal brain age gap in the severe abnormal outcome group was significantly larger than that of the normal outcome group: difference in mean brain age gap = 0.50 weeks (p-value = 0.04). CONCLUSIONS: These findings demonstrate that the deep learning model generalises to independent datasets from two clinical sites, and that the model's brain age gap magnitudes differ between neonates with normal and severe abnormal follow-up neurodevelopmental outcomes. SIGNIFICANCE: The magnitude of neonates' brain age gap, estimated using only 20 min of resting state EEG data from a single channel, can encode information of clinical neurodevelopmental value.

A ResNet mini architecture for brain age prediction.

The brain presents age-related structural and functional changes in the human life, with different extends between subjects and groups. Brain age prediction can be used to evaluate the development and aging of human brain, as well as providing valuable information for neurodevelopment and disease diagnosis. Many contributions have been made for this purpose, resorting to different machine learning methods. To solve this task and reduce memory resource consumption, we develop a mini architecture of only 10 layers by modifying the deep residual neural network (ResNet), named ResNet mini architecture. To support the ResNet mini architecture in brain age prediction, the brain age dataset (OpenNeuro #ds000228) that consists of 155 study participants (three classes) and the Alzheimer MRI preprocessed dataset that consists of 6400 images (four classes) are employed. We compared the performance of the ResNet mini architecture with other popular networks using the two considered datasets. Experimental results show that the proposed architecture exhibits generality and robustness with high accuracy and less parameter number.

The effect of head motion on brain age prediction using deep convolutional neural networks.

Deep learning can be used effectively to predict participants' age from brain magnetic resonance imaging (MRI) data, and a growing body of evidence suggests that the difference between predicted and chronological age-referred to as brain-predicted age difference (brain-PAD)-is related to various neurological and neuropsychiatric disease states. A crucial aspect of the applicability of brain-PAD as a biomarker of individual brain health is whether and how brain-predicted age is affected by MR image artifacts commonly encountered in clinical settings. To investigate this issue, we trained and validated two different 3D convolutional neural network architectures (CNNs) from scratch and tested the models on a separate dataset consisting of motion-free and motion-corrupted T1-weighted MRI scans from the same participants, the quality of which were rated by neuroradiologists from a clinical diagnostic point of view. Our results revealed a systematic increase in brain-PAD with worsening image quality for both models. This effect was also observed for images that were deemed usable from a clinical perspective, with brains appearing older in medium than in good quality images. These findings were also supported by significant associations found between the brain-PAD and standard image quality metrics indicating larger brain-PAD for lower-quality images. Our results demonstrate a spurious effect of advanced brain aging as a result of head motion and underline the importance of controlling for image quality when using brain-predicted age based on structural neuroimaging data as a proxy measure for brain health.

Brain Age Gap in Early Illness Schizophrenia and the Clinical High-Risk Syndrome: Associations With Experiential Negative Symptoms and Conversion to Psychosis.

BACKGROUND AND HYPOTHESIS: Brain development/aging is not uniform across individuals,spawning efforts to characterize brain age from a biological perspective to model the effects of disease and maladaptive life processes on the brain. The brain age gap represents the discrepancy between estimated brain biological age and chronological age (in this case, based on structural magnetic resonance imaging, MRI). Structural MRI studies report an increased brain age gap (biological age > chronological age) in schizophrenia, with a greater brain age gap related to greater negative symptom severity. Less is known regarding the nature of this gap early in schizophrenia (ESZ), if this gap represents a psychosis conversion biomarker in clinical high-risk (CHR-P) individuals, and how altered brain development and/or agingmap onto specific symptom facets. STUDY DESIGN: Using structural MRI, we compared the brain age gap among CHR-P (n = 51), ESZ (n = 78), and unaffected comparison participants (UCP; n = 90), and examined associations with CHR-P psychosis conversion (CHR-P converters n = 10; CHR-P non-converters; n = 23) and positive and negative symptoms. STUDY RESULTS: ESZ showed a greater brain age gap relative to UCP and CHR-P (Ps < .010). CHR-P individuals who converted to psychosis showed a greater brain age gap (P = .043) relative to CHR-P non-converters. A larger brain age gap in ESZ was associated with increased experiential (P = .008), but not expressive negative symptom severity. CONCLUSIONS: Consistent with schizophrenia pathophysiological models positing abnormal brain maturation, results suggest abnormal brain development is present early in psychosis. An increased brain age gap may be especially relevant to motivational and functional deficits in schizophrenia.