Predicting brain age with deep learning from raw imaging data results in a reliable and heritable biomarker

A novel network architecture combining central-peripheral deviation with image-based convolutional neural networks for diffusion tensor imaging studies

Brain imaging research is a very challenging topic due to complex structure and lack of explicitly identifiable features in the image. With the advancement of magnetic resonance imaging (MRI) technologies, such as diffusion tensor imaging (DTI), developing classification methods to improve clinical diagnosis is crucial. This paper proposes a classification method for DTI data based on a novel neural network strategy that combines a convolutional neural network (CNN) with a multilayer neural network using central-peripheral deviation (CPD), which reflects diffusion dynamics in the white matter by spatially evaluating the deviation of diffusion coefficients between the inner and outer parts of the brain. In our method, a multilayer perceptron (MLP) using CPD is combined with the final layers for classification after reducing the dimensions of images in the convolutional layers of the neural network architecture. In terms of training loss and the classification error, the proposed classification method improves the existing image classification with CNN. For real data analysis, we demonstrate how to process raw DTI image data sets obtained from a traumatic brain injury study (MagNeTS) and a brain atlas construction study (ICBM), and apply the proposed approach to the data, successfully improving classification performance with two age groups.

Connectome-based predictive models using resting-state fMRI for studying brain aging

Changes in the brain with age can provide useful information regarding an individual's chronological age. studies have suggested that functional connectomes identified via resting-state functional magnetic resonance imaging (fMRI) could be a powerful feature for predicting an individual's age. We applied connectome-based predictive modeling (CPM) to investigate individual chronological age predictions via resting-state fMRI using open-source datasets. The significant feature for age prediction was confirmed in 168 subjects from the Southwest University Adult Lifespan Dataset. The higher contributing nodes for age production included a positive connection from the left inferior parietal sulcus and a negative connection from the right middle temporal sulcus. On the network scale, the subcortical-cerebellum network was the dominant network for age prediction. The generalizability of CPM, which was constructed using the identified features, was verified by applying this model to independent datasets that were randomly selected from the Autism Brain Imaging Data Exchange I and the Open Access Series of Imaging Studies 3. CPM via resting-state fMRI is a potential robust predictor for determining an individual's chronological age from changes in the brain.

Interrelating differences in structural and functional connectivity in the older adult's brain

In the normal aging process, the functional connectome restructures and shows a shift from more segregated to more integrated brain networks, which manifests itself in highly different cognitive performances in older adults. Underpinnings of this reorganization are not fully understood, but may be related to age-related differences in structural connectivity, the underlying scaffold for information exchange between regions. The structure-function relationship might be a promising factor to understand the neurobiological sources of interindividual cognitive variability, but remain unclear in older adults. Here, we used diffusion weighted and resting-state functional magnetic resonance imaging as well as cognitive performance data of 573 older subjects from the 1000BRAINS cohort (55-85 years, 287 males) and performed a partial least square regression on 400 regional functional and structural connectivity (FC and SC, respectively) estimates comprising seven resting-state networks. Our aim was to identify FC and SC patterns that are, together with cognitive performance, characteristic of the older adults aging process. Results revealed three different aging profiles prevalent in older adults. FC was found to behave differently depending on the severity of age-related SC deteriorations. A functionally highly interconnected system is associated with a structural connectome that shows only minor age-related decreases. Because this connectivity profile was associated with the most severe age-related cognitive decline, a more interconnected FC system in older adults points to a process of dedifferentiation. Thus, functional network integration appears to increase primarily when SC begins to decline, but this does not appear to mitigate the decline in cognitive performance.

Glucose metabolism patterns: A potential index to characterize brain ageing and predict high conversion risk into cognitive impairment

Exploring individual hallmarks of brain ageing is important. Here, we propose the age-related glucose metabolism pattern (ARGMP) as a potential index to characterize brain ageing in cognitively normal (CN) elderly people. We collected 18F-fluorodeoxyglucose (18F-FDG) PET brain images from two independent cohorts: the Alzheimer's Disease Neuroimaging Initiative (ADNI, N = 127) and the Xuanwu Hospital of Capital Medical University, Beijing, China (N = 84). During follow-up (mean 80.60 months), 23 participants in the ADNI cohort converted to cognitive impairment. ARGMPs were identified using the scaled subprofile model/principal component analysis method, and cross-validations were conducted in both independent cohorts. A survival analysis was further conducted to calculate the predictive effect of conversion risk by using ARGMPs. The results showed that ARGMPs were characterized by hypometabolism with increasing age primarily in the bilateral medial superior frontal gyrus, anterior cingulate and paracingulate gyri, caudate nucleus, and left supplementary motor area and hypermetabolism in part of the left inferior cerebellum. The expression network scores of ARGMPs were significantly associated with chronological age (R = 0.808, p < 0.001), which was validated in both the ADNI and Xuanwu cohorts. Individuals with higher network scores exhibited a better predictive effect (HR: 0.30, 95% CI: 0.1340 ~ 0.6904, p = 0.0068). These findings indicate that ARGMPs derived from CN participants may represent a novel index for characterizing brain ageing and predicting high conversion risk into cognitive impairment.

Deep neural networks learn general and clinically relevant representations of the ageing brain

The discrepancy between chronological age and the apparent age of the brain based on neuroimaging data - the brain age delta - has emerged as a reliable marker of brain health. With an increasing wealth of data, approaches to tackle heterogeneity in data acquisition are vital. To this end, we compiled raw structural magnetic resonance images into one of the largest and most diverse datasets assembled (n=53542), and trained convolutional neural networks (CNNs) to predict age. We achieved state-of-the-art performance on unseen data from unknown scanners (n=2553), and showed that higher brain age delta is associated with diabetes, alcohol intake and smoking. Using transfer learning, the intermediate representations learned by our model complemented and partly outperformed brain age delta in predicting common brain disorders. Our work shows we can achieve generalizable and biologically plausible brain age predictions using CNNs trained on heterogeneous datasets, and transfer them to clinical use cases.

Multimodal MRI-Based Whole-Brain Assessment in Patients In Anoxoischemic Coma by Using 3D Convolutional Neural Networks

BACKGROUND

There is an unfulfilled need to find the best way to automatically capture, analyze, organize, and merge structural and functional brain magnetic resonance imaging (MRI) data to ultimately extract relevant signals that can assist the medical decision process at the bedside of patients in postanoxic coma. We aimed to develop and validate a deep learning model to leverage multimodal 3D MRI whole-brain times series for an early evaluation of brain damages related to anoxoischemic coma.

METHODS

This proof-of-concept, prospective, cohort study was undertaken at the intensive care unit affiliated with the University Hospital (Toulouse, France), between March 2018 and May 2020. All patients were scanned in coma state at least 2 days (4 ± 2 days) after cardiac arrest. Over the same period, age-matched healthy volunteers were recruited and included. Brain MRI quantification encompassed both "functional data" from regions of interest (precuneus and posterior cingulate cortex) with whole-brain functional connectivity analysis and "structural data" (gray matter volume, T1-weighted, fractional anisotropy, and mean diffusivity). A specifically designed 3D convolutional neuronal network (CNN) was created to allow conscious state discrimination (coma vs. controls) by using raw MRI indices as the input. A voxel-wise visualization method based on the study of convolutional filters was applied to support CNN outcome. The Ethics Committee of the University Teaching Hospital of Toulouse, France (2018-A31) approved the study and informed consent was obtained from all participants.

RESULTS

The final cohort consisted of 29 patients in postanoxic coma and 34 healthy volunteers. Coma patients were successfully discerned from controls by using 3D CNN in combination with different MR indices. The best accuracy was achieved by functional MRI data, in particular with resting-state functional MRI of the posterior cingulate cortex, with an accuracy of 0.96 (range 0.94-0.98) on the test set from 10-time repeated tenfold cross-validation. Even more satisfactory performances were achieved through the majority voting strategy, which was able to compensate for mistakes from single MR indices. Visualization maps allowed us to identify the most relevant regions for each MRI index, notably regions previously described as possibly being involved in consciousness emergence. Interestingly, a posteriori analysis of misclassified patients indicated that they may present some common functional MRI traits with controls, which suggests further favorable outcomes.

CONCLUSIONS

A fully automated identification of clinically relevant signals from complex multimodal neuroimaging data is a major research topic that may bring a radical paradigm shift in the neuroprognostication of patients with severe brain injury. We report for the first time a successful discrimination between patients in postanoxic coma patients from people serving as controls by using 3D CNN whole-brain structural and functional MRI data. Clinical Trial Number http://ClinicalTrials.gov (No. NCT03482115).

A reusable benchmark of brain-age prediction from M/EEG resting-state signals

Population-level modeling can define quantitative measures of individual aging by applying machine learning to large volumes of brain images. These measures of brain age, obtained from the general population, helped characterize disease severity in neurological populations, improving estimates of diagnosis or prognosis. Magnetoencephalography (MEG) and Electroencephalography (EEG) have the potential to further generalize this approach towards prevention and public health by enabling assessments of brain health at large scales in socioeconomically diverse environments. However, more research is needed to define methods that can handle the complexity and diversity of M/EEG signals across diverse real-world contexts. To catalyse this effort, here we propose reusable benchmarks of competing machine learning approaches for brain age modeling. We benchmarked popular classical machine learning pipelines and deep learning architectures previously used for pathology decoding or brain age estimation in 4 international M/EEG cohorts from diverse countries and cultural contexts, including recordings from more than 2500 participants. Our benchmarks were built on top of the M/EEG adaptations of the BIDS standard, providing tools that can be applied with minimal modification on any M/EEG dataset provided in the BIDS format. Our results suggest that, regardless of whether classical machine learning or deep learning was used, the highest performance was reached by pipelines and architectures involving spatially aware representations of the M/EEG signals, leading to R^2 scores between 0.60-0.71. Hand-crafted features paired with random forest regression provided robust benchmarks even in situations in which other approaches failed. Taken together, this set of benchmarks, accompanied by open-source software and high-level Python scripts, can serve as a starting point and quantitative reference for future efforts at developing M/EEG-based measures of brain aging. The generality of the approach renders this benchmark reusable for other related objectives such as modeling specific cognitive variables or clinical endpoints.

Towards the Interpretability of Deep Learning Models for Multi-modal Neuroimaging: Finding Structural Changes of the Ageing Brain

Brain-age (BA) estimates based on deep learning are increasingly used as neuroimaging biomarker for brain health; however, the underlying neural features have remained unclear. We combined ensembles of convolutional neural networks with Layer-wise Relevance Propagation (LRP) to detect which brain features contribute to BA. Trained on magnetic resonance imaging (MRI) data of a population-based study (n=2637, 18-82 years), our models estimated age accurately based on single and multiple modalities, regionally restricted and whole-brain images (mean absolute errors 3.37-3.86 years). We find that BA estimates capture aging at both small and large-scale changes, revealing gross enlargements of ventricles and subarachnoid spaces, as well as white matter lesions, and atrophies that appear throughout the brain. Divergence from expected aging reflected cardiovascular risk factors and accelerated aging was more pronounced in the frontal lobe. Applying LRP, our study demonstrates how superior deep learning models detect brain-aging in healthy and at-risk individuals throughout adulthood.

Robust Brain Age Estimation based on sMRI via Nonlinear Age-Adaptive Ensemble Learning

Precise prediction on brain age is urgently needed by many biomedical areas including mental rehabilitation prognosis as well as various medicine or treatment trials. People began to realize that contrasting physical (real) age and predicted brain age can help to highlight brain issues and evaluate if patients' brains are healthy or not. Such age prediction is often challenging for single model-based prediction, while the conditions of brains vary drastically over age. In this work, we present an age-adaptive ensemble model that is based on the combination of four different machine learning algorithms, including a support vector machine (SVR), a convolutional neural network (CNN) model, and the popular GoogLeNet and ResNet deep networks. The ensemble model proposed here is nonlinearly adaptive, where age is taken as a key factor in the nonlinear combination of various single-algorithm-based independent models. In our age-adaptive ensemble method, the weights of each model are learned automatically as nonlinear functions over age instead of fixed values, while brain age estimation is based on such an age-adaptive integration of various single models. The quality of the model is quantified by the mean absolute errors (MAE) and spearman correlation between the predicted age and the actual age, with the least MAE and the highest Spearman correlation representing the highest accuracy in age prediction. By testing on the Predictive Analysis Challenge 2019 (PAC 2019) dataset, our novel ensemble model has achieved a MAE down to 3.19, which is a significantly increased accuracy in this brain age competition. If deployed in the real world, our novel ensemble model having an improved accuracy could potentially help doctors to identify the risk of brain diseases more accurately and quickly, thus helping pharmaceutical companies develop drugs or treatments precisely, and potential offer a new powerful tool for researchers in the field of brain science.

Identification of Children's Tuberous Sclerosis Complex with Multiple-contrast MRI and 3D Convolutional Network

Identifying rare tuberous sclerosis complex (TSC) children is valuable and crucial. Magnetic resonance imaging (MRI) is used for rare TSC diagnoses. In this work, T2w and FLAIR were combined as a new modality named FLAIR3 to maximize the contrast between TSC lesions and normal-appearing brain tissues. After that, for the first time, we propose to use two different 3D CNN combined with late fusion strategies to diagnose TSC. A total of 520 children were enrolled in the study, including 260 health and 260 TSC children. The experiments had shown that the FLAIR3 could effectively improve the conspicuity of TSC lesions and classification performance. And the results showed the proposed late fusion method can improve the classification performance and achieve the state-of-the-art performance of the AUC of 0.994 and the accuracy of 0.971, which could be treated as an effective computer-aided diagnostic tool to help clinical radiologists diagnose TSC children. Clinical Relevance- Our deep learning method can be a non-invasive, efficient, and reliable way to help clinical radiologists to identify TSC patients. FLAIR3 can provide clinicians with a new modality to accurately localize TSC lesions in TSC patients.

Multimodal MRI examination of structural and functional brain changes in older women with breast cancer in the first year of antiestrogen hormonal therapy

PURPOSE

Cancer patients are concerned about treatment-related cognitive problems. We examined effects of antiestrogen hormonal therapy on brain imaging metrics in older women with breast cancer.

METHODS

Women aged 60 + treated with hormonal therapy only and matched non-cancer controls (n = 29/group) completed MRI and objective and self-reported cognitive assessment at pre-treatment/enrollment and 12 months later. Gray matter was examined using voxel-based morphometry (VBM), FreeSurfer, and brain age calculations. Functional MRI (fMRI) assessed working memory-related activation. Analyses examined cross-sectional and longitudinal differences and tested associations between brain metrics, cognition, and days on hormonal therapy.

RESULTS

The cancer group showed regional reductions over 12 months in frontal, temporal, and parietal gray matter on VBM, reduced FreeSurfer cortical thickness in prefrontal, parietal, and insular regions, and increased working memory-related fMRI activation in frontal, cingulate, and visual association cortex. Controls showed only reductions in fusiform gyrus on VBM and FreeSurfer temporal and parietal cortex thickness. Women with breast cancer showed higher estimated brain age and lower regional gray matter volume than controls at both time points. The cancer group showed a trend toward lower performance in attention, processing speed, and executive function at follow-up. There were no significant associations between brain imaging metrics and cognition or days on hormonal therapy.

CONCLUSION

Older women with breast cancer showed brain changes in the first year of hormonal therapy. Increased brain activation during working memory processing may be a sign of functional compensation for treatment-related structural changes. This hypothesis should be tested in larger samples over longer time periods.

CLINICALTRIALS

GOV IDENTIFIER

NCT03451383.

Probing the link between the APOE-ε4 allele and whole-brain gray matter using deep learning

The APOE-ε4 allele is a known genetic risk for Alzheimer's disease (AD). Thus, it can be reasoned that the APOE-ε4 allele would also impact neurodegeneration-associated structural brain changes. Here we probe if the APOE-ε4 genotype directly modulates the human brain's gray matter using a neural network trained on the whole-brain gray matter images from the cognitively normally aging (CN) and AD individuals. To investigate the linkage between the APOE-ε4 allele and whole-brain (voxel-wise) gray matter, we systematically profile our investigation in multiple classification tasks, including diagnostic classification and APOE-ε4 classification conjointly as well as independently. Results suggest that although the MRI data can reliably track and reflect neurodegenerative changes in the brain cross-sectionally, the APOE-ε4 status may not be distinguishable correspondingly. The nonexistence of a direct and convincing modulative effect of APOE-ε4 on the whole-brain gray matter indicates that the gray matter changes may be independent of the APOE-ε4 status, and instead characterize a non-APOE, comorbid mechanism in AD.

Mind the gap: Performance metric evaluation in brain-age prediction

Estimating age based on neuroimaging-derived data has become a popular approach to developing markers for brain integrity and health. While a variety of machine-learning algorithms can provide accurate predictions of age based on brain characteristics, there is significant variation in model accuracy reported across studies. We predicted age in two population-based datasets, and assessed the effects of age range, sample size and age-bias correction on the model performance metrics Pearson's correlation coefficient (r), the coefficient of determination (R2 ), Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE). The results showed that these metrics vary considerably depending on cohort age range; r and R2 values are lower when measured in samples with a narrower age range. RMSE and MAE are also lower in samples with a narrower age range due to smaller errors/brain age delta values when predictions are closer to the mean age of the group. Across subsets with different age ranges, performance metrics improve with increasing sample size. Performance metrics further vary depending on prediction variance as well as mean age difference between training and test sets, and age-bias corrected metrics indicate high accuracy-also for models showing poor initial performance. In conclusion, performance metrics used for evaluating age prediction models depend on cohort and study-specific data characteristics, and cannot be directly compared across different studies. Since age-bias corrected metrics generally indicate high accuracy, even for poorly performing models, inspection of uncorrected model results provides important information about underlying model attributes such as prediction variance.

Impact of Weight Loss on Brain Age: Improved Brain Health Following Bariatric Surgery

Individuals living with obesity tend to have increased brain age, reflecting poorer brain health likely due to grey and white matter atrophy related to obesity. However, it is unclear if older brain age associated with obesity can be reversed following weight loss and cardiometabolic health improvement. The aim of this study was to assess the impact of weight loss and cardiometabolic improvement following bariatric surgery on brain health, as measured by change in brain age estimated based on voxel-based morphometry (VBM) measurements. We used three distinct datasets to perform this study: 1) CamCAN dataset to train the brain age prediction model, 2) Human Connectome Project (HCP) dataset to investigate whether individuals with obesity have greater brain age than individuals with normal weight, and 3) pre-surgery, as well as 4, 12, and 24 month post-surgery data from participants (n=87, age: 44.0±9.2 years, BMI: 43.9±4.2 kg/m2) who underwent a bariatric surgery to investigate whether weight loss and cardiometabolic improvement as a result of bariatric surgery lowers the brain age. As expected, our results from the HCP dataset showed a higher brain age for individuals with obesity compared to individuals with normal weight (T-value = 7.08, p-value < 0.0001). We also found significant improvement in brain health, indicated by a decrease of 2.9 and 5.6 years in adjusted delta age at 12 and 24 months following bariatric surgery compared to baseline (p-value < 0.0005 for both). While the overall effect seemed to be driven by a global change across all brain regions and not from a specific region, our exploratory analysis showed lower delta age in certain brain regions (mainly in somatomotor, visual, and ventral attention networks) at 24 months. This reduced age was also associated with post-surgery improvements in BMI, systolic/diastolic blood pressure, and HOMA-IR (T-value_BMI=4.29, T-value_SBP=4.67, T-value_DBP=4.12, T-value_HOMA-IR=3.16, all p-values < 0.05). In conclusion, these results suggest that obesity-related brain health abnormalities (as measured by delta age) might be reversed by bariatric surgery-induced weight loss and widespread improvements in cardiometabolic alterations.

Disentangling Alzheimer's disease neurodegeneration from typical brain ageing using machine learning

Neuroimaging biomarkers that distinguish between changes due to typical brain ageing and Alzheimer's disease are valuable for determining how much each contributes to cognitive decline. Supervised machine learning models can derive multivariate patterns of brain change related to the two processes, including the Spatial Patterns of Atrophy for Recognition of Alzheimer's Disease (SPARE-AD) and of Brain Aging (SPARE-BA) scores investigated herein. However, the substantial overlap between brain regions affected in the two processes confounds measuring them independently. We present a methodology, and associated results, towards disentangling the two. T1-weighted MRI scans of 4054 participants (48-95 years) with Alzheimer's disease, mild cognitive impairment (MCI), or cognitively normal (CN) diagnoses from the Imaging-based coordinate SysTem for AGIng and NeurodeGenerative diseases (iSTAGING) consortium were analysed. Multiple sets of SPARE scores were investigated, in order to probe imaging signatures of certain clinically or molecularly defined sub-cohorts. First, a subset of clinical Alzheimer's disease patients (n = 718) and age- and sex-matched CN adults (n = 718) were selected based purely on clinical diagnoses to train SPARE-BA1 (regression of age using CN individuals) and SPARE-AD1 (classification of CN versus Alzheimer's disease) models. Second, analogous groups were selected based on clinical and molecular markers to train SPARE-BA2 and SPARE-AD2 models: amyloid-positive Alzheimer's disease continuum group (n = 718; consisting of amyloid-positive Alzheimer's disease, amyloid-positive MCI, amyloid- and tau-positive CN individuals) and amyloid-negative CN group (n = 718). Finally, the combined group of the Alzheimer's disease continuum and amyloid-negative CN individuals was used to train SPARE-BA3 model, with the intention to estimate brain age regardless of Alzheimer's disease-related brain changes. The disentangled SPARE models, SPARE-AD2 and SPARE-BA3, derived brain patterns that were more specific to the two types of brain changes. The correlation between the SPARE-BA Gap (SPARE-BA minus chronological age) and SPARE-AD was significantly reduced after the decoupling (r = 0.56-0.06). The correlation of disentangled SPARE-AD was non-inferior to amyloid- and tau-related measurements and to the number of APOE ε4 alleles but was lower to Alzheimer's disease-related psychometric test scores, suggesting the contribution of advanced brain ageing to the latter. The disentangled SPARE-BA was consistently less correlated with Alzheimer's disease-related clinical, molecular and genetic variables. By employing conservative molecular diagnoses and introducing Alzheimer's disease continuum cases to the SPARE-BA model training, we achieved more dissociable neuroanatomical biomarkers of typical brain ageing and Alzheimer's disease.

Artificial intelligence in brain MRI analysis of Alzheimer's disease over the past 12 years: A systematic review

INTRODUCTION

Multiple structural brain changes in Alzheimer's disease (AD) and mild cognitive impairment (MCI) have been revealed on magnetic resonance imaging (MRI). There is a fast-growing effort in applying artificial intelligence (AI) to analyze these data. Here, we review and evaluate the AI studies in brain MRI analysis with synthesis.

METHODS

A systematic review of the literature, spanning the years from 2009 to 2020, was completed using the PubMed database. AI studies using MRI imaging to investigate normal aging, mild cognitive impairment, and AD-dementia were retrieved for review. Bias assessment was completed using the PROBAST criteria.

RESULTS

97 relevant studies were included in the review. The studies were typically focused on the classification of AD, MCI, and normal aging (71% of the reported studies) and the prediction of MCI conversion to AD (25%). The best performance was achieved by using the deep learning-based convolution neural network algorithms (weighted average accuracy 89%), in contrast to 76-86% using Logistic Regression, Support Vector Machines, and other AI methods.

DISCUSSION

The synthesized evidence is paramount to developing sophisticated AI approaches to reliably capture and quantify multiple subtle MRI changes in the whole brain that exemplify the complexity and heterogeneity of AD and brain aging.

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

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. Here, we developed a deep learning-based brain age prediction model using a large collection of fluorodeoxyglucose positron emission tomography and structural magnetic resonance imaging and tested how the brain age gap relates to degenerative syndromes including mild cognitive impairment, AD, frontotemporal dementia and Lewy body dementia. Occlusion analysis, performed to facilitate the interpretation of the model, revealed that the model learns an age- and modality-specific pattern of brain aging. The elevated brain age gap was highly correlated with cognitive impairment and the AD biomarker. The higher gap also showed a longitudinal predictive nature across clinical categories, including cognitively unimpaired individuals who converted to a clinical stage. However, regions generating brain age gaps were different for each diagnostic group of which the AD continuum showed similar patterns to normal aging.

Prediction of Chronological Age in Healthy Elderly Subjects with Machine Learning from MRI Brain Segmentation and Cortical Parcellation

Normal aging is associated with changes in volumetric indices of brain atrophy. A quantitative understanding of age-related brain changes can shed light on successful aging. To investigate the effect of age on global and regional brain volumes and cortical thickness, 3514 magnetic resonance imaging scans were analyzed using automated brain segmentation and parcellation methods in elderly healthy individuals (69–88 years of age). The machine learning algorithm extreme gradient boosting (XGBoost) achieved a mean absolute error of 2 years in predicting the age of new subjects. Feature importance analysis showed that the brain-to-intracranial-volume ratio is the most important feature in predicting age, followed by the hippocampi volumes. The cortical thickness in temporal and parietal lobes showed a superior predictive value than frontal and occipital lobes. Insights from this approach that integrate model prediction and interpretation may help to shorten the current explanatory gap between chronological age and biological brain age.

Machine Learning Quantifies Accelerated White-Matter Aging in Persons With HIV

BACKGROUND

Persons with HIV (PWH) undergo white matter changes, which can be quantified using the brain-age gap (BAG), the difference between chronological age and neuroimaging-based brain-predicted age. Accumulation of microstructural damage may be accelerated in PWH, especially with detectable viral load (VL).

METHODS

In total, 290 PWH (85% with undetectable VL) and 165 HIV-negative controls participated in neuroimaging and cognitive testing. BAG was measured using a Gaussian process regression model trained to predict age from diffusion magnetic resonance imaging in publicly available normative controls. To test for accelerated aging, BAG was modeled as an age × VL interaction. The relationship between BAG and global neuropsychological performance was examined. Other potential predictors of pathological aging were investigated in an exploratory analysis.

RESULTS

Age and detectable VL had a significant interactive effect: PWH with detectable VL accumulated +1.5 years BAG/decade versus HIV-negative controls (P = .018). PWH with undetectable VL accumulated +0.86 years BAG/decade, although this did not reach statistical significance (P = .052). BAG was associated with poorer global cognition only in PWH with detectable VL (P < .001). Exploratory analysis identified Framingham cardiovascular risk as an additional predictor of pathological aging (P = .027).

CONCLUSIONS

Aging with detectable HIV and cardiovascular disease may lead to white matter pathology and contribute to cognitive impairment.

Structural and Functional Trajectories of Middle Temporal Gyrus Sub-Regions During Life Span: A Potential Biomarker of Brain Development and Aging

Although previous studies identified a similar topography pattern of structural and functional delineations in human middle temporal gyrus (MTG) using healthy adults, trajectories of MTG sub-regions across lifespan remain largely unknown. Herein, we examined gray matter volume (GMV) and resting-state functional connectivity (RSFC) using datasets from the Nathan Kline Institute (NKI), and aimed to (1) investigate structural and functional trajectories of MTG sub-regions across the lifespan; and (2) assess whether these features can be used as biomarkers to predict individual’s chronological age. As a result, GMV of all MTG sub-regions followed U-shaped trajectories with extreme age around the sixth decade. The RSFC between MTG sub-regions and many cortical brain regions showed inversed U-shaped trajectories, whereas RSFC between MTG sub-regions and sub-cortical regions/cerebellum showed U-shaped way, with extreme age about 20 years earlier than those of GMV. Moreover, GMV and RSFC of MTG sub-regions could be served as useful features to predict individual age with high estimation accuracy. Together, these results not only provided novel insights into the dynamic process of structural and functional roles of MTG sub-regions across the lifespan, but also served as useful biomarkers to age prediction.

Semi-Synchronous Federated Learning for Energy-Efficient Training and Accelerated Convergence in Cross-Silo Settings

There are situations where data relevant to machine learning problems are distributed across multiple locations that cannot share the data due to regulatory, competitiveness, or privacy reasons. Machine learning approaches that require data to be copied to a single location are hampered by the challenges of data sharing. Federated Learning (FL) is a promising approach to learn a joint model over all the available data across silos. In many cases, the sites participating in a federation have different data distributions and computational capabilities. In these heterogeneous environments existing approaches exhibit poor performance: synchronous FL protocols are communication efficient, but have slow learning convergence and high energy cost; conversely, asynchronous FL protocols have faster convergence with lower energy cost, but higher communication. In this work, we introduce a novel energy-efficient Semi-Synchronous Federated Learning protocol that mixes local models periodically with minimal idle time and fast convergence. We show through extensive experiments over established benchmark datasets in the computer-vision domain as well as in real-world biomedical settings that our approach significantly outperforms previous work in data and computationally heterogeneous environments .

Advanced brain age in deployment-related traumatic brain injury: A LIMBIC-CENC neuroimaging study

OBJECTIVE

To determine if history of mild traumatic brain injury (mTBI) is associated with advanced or accelerated brain aging among the United States (US) military Service Members and Veterans.

METHODS

Eight hundred and twenty-two participants (mean age = 40.4 years, 714 male/108 female) underwent MRI sessions at eight sites across the US. Two hundred and one participants completed a follow-up scan between five months and four years later. Predicted brain ages were calculated using T1-weighted MRIs and then compared with chronological ages to generate an Age Deviation Score for cross-sectional analyses and an Interval Deviation Score for longitudinal analyses. Participants also completed a neuropsychological battery, including measures of both cognitive functioning and psychological health.

RESULT

In cross-sectional analyses, males with a history of deployment-related mTBI showed advanced brain age compared to those without (t(884) = 2.1, p = .038), while this association was not significant in females. In follow-up analyses of the male participants, severity of posttraumatic stress disorder (PTSD), depression symptoms, and alcohol misuse were also associated with advanced brain age.

CONCLUSION

History of deployment-related mTBI, severity of PTSD and depression symptoms, and alcohol misuse are associated with advanced brain aging in male US military Service Members and Veterans.

The Impact of Genes and Environment on Brain Ageing in Males Aged 51 to 72 Years

Magnetic resonance imaging data are being used in statistical models to predicted brain ageing (PBA) and as biomarkers for neurodegenerative diseases such as Alzheimer's Disease. Despite their increasing application, the genetic and environmental etiology of global PBA indices is unknown. Likewise, the degree to which genetic influences in PBA are longitudinally stable and how PBA changes over time are also unknown. We analyzed data from 734 men from the Vietnam Era Twin Study of Aging with repeated MRI assessments between the ages 51-72 years. Biometrical genetic analyses "twin models" revealed significant and highly correlated estimates of additive genetic heritability ranging from 59 to 75%. Multivariate longitudinal modeling revealed that covariation between PBA at different timepoints could be explained by a single latent factor with 73% heritability. Our results suggest that genetic influences on PBA are detectable in midlife or earlier, are longitudinally very stable, and are largely explained by common genetic influences.

The impact of Alzheimer's disease susceptibility loci on lateral ventricular surface morphology in older adults

The enlargement of ventricular volume is a general trend in the elderly, especially in patients with Alzheimer's disease (AD). Multiple susceptibility loci have been reported to have an increased risk for AD and the morphology of brain structures are affected by the variations in the risk loci. Therefore, we hypothesized that genes contributed significantly to the ventricular surface, and the changes of ventricular surface were associated with the impairment of cognitive functions. After the quality controls (QC) and genotyping, a lateral ventricular segmentation method was employed to obtain the surface features of lateral ventricle. We evaluated the influence of 18 selected AD susceptibility loci on both volume and surface morphology across 410 subjects from Alzheimer's Disease Neuroimaging Initiative (ADNI). Correlations were conducted between radial distance (RD) and Montreal Cognitive Assessment (MoCA) subscales. Only the C allele at the rs744373 loci in BIN1 gene significantly accelerated the atrophy of lateral ventricle, including the anterior horn, body, and temporal horn of left lateral ventricle. No significant effect on lateral ventricle was found at other loci. Our results revealed that most regions of the bilateral ventricular surface were significantly negatively correlated with cognitive scores, particularly in delayed recall. Besides, small areas of surface were negatively correlated with language, orientation, and visuospatial scores. Together, our results indicated that the genetic variation affected the localized areas of lateral ventricular surface, and supported that lateral ventricle was an important brain structure associated with cognition in the elderly.

Accurate brain-age models for routine clinical MRI examinations

Convolutional neural networks (CNN) can accurately predict chronological age in healthy individuals from structural MRI brain scans. Potentially, these models could be applied during routine clinical examinations to detect deviations from healthy ageing, including early-stage neurodegeneration. This could have important implications for patient care, drug development, and optimising MRI data collection. However, existing brain-age models are typically optimised for scans which are not part of routine examinations (e.g., volumetric T1-weighted scans), generalise poorly (e.g., to data from different scanner vendors and hospitals etc.), or rely on computationally expensive pre-processing steps which limit real-time clinical utility. Here, we sought to develop a brain-age framework suitable for use during routine clinical head MRI examinations. Using a deep learning-based neuroradiology report classifier, we generated a dataset of 23,302 'radiologically normal for age' head MRI examinations from two large UK hospitals for model training and testing (age range = 18-95 years), and demonstrate fast (< 5 s), accurate (mean absolute error [MAE] < 4 years) age prediction from clinical-grade, minimally processed axial T2-weighted and axial diffusion-weighted scans, with generalisability between hospitals and scanner vendors (Δ MAE < 1 year). The clinical relevance of these brain-age predictions was tested using 228 patients whose MRIs were reported independently by neuroradiologists as showing atrophy 'excessive for age'. These patients had systematically higher brain-predicted age than chronological age (mean predicted age difference = +5.89 years, 'radiologically normal for age' mean predicted age difference = +0.05 years, p < 0.0001). Our brain-age framework demonstrates feasibility for use as a screening tool during routine hospital examinations to automatically detect older-appearing brains in real-time, with relevance for clinical decision-making and optimising patient pathways.

Brain age estimation using multi-feature-based networks

Studying brain aging improves our understanding in differentiating typical and atypical aging. Directly utilizing traditional morphological features for brain age estimation did not show significant performance in healthy controls (HCs), which may be due to the negligence of the information of structural similarities among cortical regions. For this issue, the multi-feature-based network (MFN) built upon morphological features can be employed to describe these similarities. Based on this, we hypothesized that the MFN is more efficient and robust than traditional morphological features in brain age estimating. In this work, we used six different types of morphological features (i.e., cortical volume, cortical thickness, curvature index, folding index, local gyrification index, and surface area) to build individual MFN for brain age estimation. The efficacy of MFN was estimated on 2501 HCs with T1-weighted structural magnetic resonance imaging (sMRI) data and compared with traditional morphological features. We attained a mean absolute error (MAE) of 3.73 years using the proposed method on an independent test set, whereas a mean absolute error of 5.30 years was derived from morphological features. Our experimental results demonstrated that the MFN is an efficient and robust metric for estimating brain age.

Ten years of image analysis and machine learning competitions in dementia

Machine learning methods exploiting multi-parametric biomarkers, especially based on neuroimaging, have huge potential to improve early diagnosis of dementia and to predict which individuals are at-risk of developing dementia. To benchmark algorithms in the field of machine learning and neuroimaging in dementia and assess their potential for use in clinical practice and clinical trials, seven grand challenges have been organized in the last decade: MIRIAD (2012), Alzheimer's Disease Big Data DREAM (2014), CADDementia (2014), Machine Learning Challenge (2014), MCI Neuroimaging (2017), TADPOLE (2017), and the Predictive Analytics Competition (2019). Based on two challenge evaluation frameworks, we analyzed how these grand challenges are complementing each other regarding research questions, datasets, validation approaches, results and impact. The seven grand challenges addressed questions related to screening, clinical status estimation, prediction and monitoring in (pre-clinical) dementia. There was little overlap in clinical questions, tasks and performance metrics. Whereas this aids providing insight on a broad range of questions, it also limits the validation of results across challenges. The validation process itself was mostly comparable between challenges, using similar methods for ensuring objective comparison, uncertainty estimation and statistical testing. In general, winning algorithms performed rigorous data pre-processing and combined a wide range of input features. Despite high state-of-the-art performances, most of the methods evaluated by the challenges are not clinically used. To increase impact, future challenges could pay more attention to statistical analysis of which factors (i.e., features, models) relate to higher performance, to clinical questions beyond Alzheimer's disease, and to using testing data beyond the Alzheimer's Disease Neuroimaging Initiative. Grand challenges would be an ideal venue for assessing the generalizability of algorithm performance to unseen data of other cohorts. Key for increasing impact in this way are larger testing data sizes, which could be reached by sharing algorithms rather than data to exploit data that cannot be shared. Given the potential and lessons learned in the past ten years, we are excited by the prospects of grand challenges in machine learning and neuroimaging for the next ten years and beyond.

Quantification of brain age using high-resolution 7 tesla MR imaging and implications for patients with epilepsy

Purpose

Epilepsy patients exhibit morphological differences on neuroimaging compared to age-matched healthy controls, including cortical and sub-cortical volume loss and altered gray-white matter ratios. The objective was to develop a model of normal aging using the 7T MRIs of healthy controls. This model can then be used to determine if the changes in epilepsy patients resemble the changes seen in aging, and potentially give a marker for the severity of those changes.

Methods

Sixty-nine healthy controls (24F/45M, mean age 36.5 ± 10.5 years) and forty-four epilepsy patients (24F/20M, 33.2 ± 9.9 years) non-lesional at 3T were scanned with volumetric T1-MPRAGE at 7T. These images were segmented and quantified using FreeSurfer. A linear regression-based model trained on healthy controls was developed to predict ages using derived imaging features among the epilepsy patient cohort. The model used 114 features with significant linear correlation with age.

Results

The regression-based model estimated brain age with mean absolute error (MAE) of 6.6 years among controls. Comparable prediction accuracy of 6.9 years MAE was seen epilepsy patients. T-test of mean absolute error showed no difference in the prediction accuracy with controls and epilepsy patients (p = 0.68). However, average signed error showed elevated (+5.0 years, p = 0.0007) predicted age differences (PAD; brain-PAD=, predicted minus biological age) among epilepsy patients. Morphological metrics in the medial temporal lobe were major contributors to PAD. Additionally, patients with seizure frequency greater than once a week showed significantly elevated brain-PAD (+8.2 ± 5.3 years, n = 13) compared to patients with lower seizure frequency (3.7 ± 6.5 years, n = 31, p = 0.033).

Major conclusions

Morphological patterns suggestive of premature aging were observed in non-lesional epilepsy patients vs. controls and in high seizure frequency patients vs. low frequency patients. Modeling brain age with 7T MRI may provide a sensitive imaging marker to assess the differential effects of the aging process in diseases such as epilepsy.

Brain-age estimation accuracy is significantly increased using multishell free-water reconstruction

Although free-water diffusion reconstruction for diffusion-weighted imaging (DWI) data can be applied to both single-shell and multishell data, recent finding in synthetic data suggests that the free-water indices from single-shell acquisition should be interpreted with care, as they are heavily influenced by initialization parameters and cannot discriminate between free-water and mean diffusivity modifications. However, whether using a longer multishell acquisition protocol significantly improve reconstruction for real human MRI data is still an open question. In this study, we compare canonical diffusion tensor imaging (DTI), single-shell and multishell free-water imaging (FW) indices derived from a short, clinical compatible diffusion protocol (b = 500 s/mm² , b = 1,000 s/mm² , 32 directions each) on their power to predict brain age. Age was chosen as it is well-known to be related to widespread modification of the white matter and because brain-age estimation has recently been found to be relevant to several neurodegenerative diseases. We used a previously developed and validated data-driven whole-brain machine learning pipeline to directly compare the precision of brain-age estimates in a sample of 89 healthy males between 20 and 85 years old. We found that multishell FW outperform DTI indices in estimating brain age and that multishell FW, even when using low (500 ms² ) b-values secondary shell, outperform single-shell FW. Single-shell FW led to lower brain-age estimation accuracy even of canonical DTI indices, suggesting that single-shell FW indices should be used with caution. For all considered reconstruction algorithms, the most discriminant indices were those measuring free diffusion of water in the white matter.

Multimodal biological brain age prediction using magnetic resonance imaging and angiography with the identification of predictive regions

Biological brain age predicted using machine learning models based on high-resolution imaging data has been suggested as a potential biomarker for neurological and cerebrovascular diseases. In this work, we aimed to develop deep learning models to predict the biological brain age using structural magnetic resonance imaging and angiography datasets from a large database of 2074 adults (21-81 years). Since different imaging modalities can provide complementary information, combining them might allow to identify more complex aging patterns, with angiography data, for instance, showing vascular aging effects complementary to the atrophic brain tissue changes seen in T1-weighted MRI sequences. We used saliency maps to investigate the contribution of cortical, subcortical, and arterial structures to the prediction. Our results show that combining T1-weighted and angiography MR data led to a significantly improved brain age prediction accuracy, with a mean absolute error of 3.85 years comparing the predicted and chronological age. The most predictive brain regions included the lateral sulcus, the fourth ventricle, and the amygdala, while the brain arteries contributing the most to the prediction included the basilar artery, the middle cerebral artery M2 segments, and the left posterior cerebral artery. Our study proposes a framework for brain age prediction using multimodal imaging, which gives accurate predictions and allows identifying the most predictive regions for this task, which can serve as a surrogate for the brain regions that are most affected by aging.

Domain- and task-specific transfer learning for medical segmentation tasks

BACKGROUND AND OBJECTIVES

Transfer learning is a valuable approach to perform medical image segmentation in settings with limited cases available for training convolutional neural networks (CNN). Both the source task and the source domain influence transfer learning performance on a given target medical image segmentation task. This study aims to assess transfer learning-based medical segmentation task performance for various source task and domain combinations.

METHODS

CNNs were pre-trained on classification, segmentation, and self-supervised tasks on two domains: natural images and T1 brain MRI. Next, these CNNs were fine-tuned on three target T1 brain MRI segmentation tasks: stroke lesion, MS lesions, and brain anatomy segmentation. In all experiments, the CNN architecture and transfer learning strategy were the same. The segmentation accuracy on all target tasks was evaluated using the mIOU or Dice coefficients. The detection accuracy was evaluated for the stroke and MS lesion target tasks only.

RESULTS

CNNs pre-trained on a segmentation task on the same domain as the target tasks resulted in higher or similar segmentation accuracy compared to other source task and domain combinations. Pre-training a CNN on ImageNet resulted in a comparable, but not consistently higher lesion detection rate, despite the amount of training data used being 10 times larger.

CONCLUSIONS

This study suggests that optimal transfer learning for medical segmentation is achieved with a similar task and domain for pre-training. As a result, CNNs can be effectively pre-trained on smaller datasets by selecting a source domain and task similar to the target domain and task.

Cardiometabolic risk factors associated with brain age and accelerate brain ageing

The structure and integrity of the ageing brain is interchangeably linked to physical health, and cardiometabolic risk factors (CMRs) are associated with dementia and other brain disorders. In this mixed cross-sectional and longitudinal study (interval mean = 19.7 months), including 790 healthy individuals (mean age = 46.7 years, 53% women), we investigated CMRs and health indicators including anthropometric measures, lifestyle factors, and blood biomarkers in relation to brain structure using MRI-based morphometry and diffusion tensor imaging (DTI). We performed tissue specific brain age prediction using machine learning and performed Bayesian multilevel modeling to assess changes in each CMR over time, their respective association with brain age gap (BAG), and their interaction effects with time and age on the tissue-specific BAGs. The results showed credible associations between DTI-based BAG and blood levels of phosphate and mean cell volume (MCV), and between T1-based BAG and systolic blood pressure, smoking, pulse, and C-reactive protein (CRP), indicating older-appearing brains in people with higher cardiometabolic risk (smoking, higher blood pressure and pulse, low-grade inflammation). Longitudinal evidence supported interactions between both BAGs and waist-to-hip ratio (WHR), and between DTI-based BAG and systolic blood pressure and smoking, indicating accelerated ageing in people with higher cardiometabolic risk (smoking, higher blood pressure, and WHR). The results demonstrate that cardiometabolic risk factors are associated with brain ageing. While randomized controlled trials are needed to establish causality, our results indicate that public health initiatives and treatment strategies targeting modifiable cardiometabolic risk factors may also improve risk trajectories and delay brain ageing.

Harmonization of multi-site functional connectivity measures in tangent space improves brain age prediction

Brain age prediction based on functional magnetic resonance imaging (fMRI) data has the potential to serve as a biomarker for quantifying brain health. To predict the brain age based on fMRI data robustly and accurately, we curated a large dataset (n = 4259) of fMRI scans from seven different data acquisition sites and computed personalized functional connectivity measures at multiple scales from each subject's fMRI scan. Particularly, we computed personalized large-scale functional networks and generated functional connectivity measures at multiple scales to characterize each fMRI scan. To account for inter-site effects on the functional connectivity measures, we harmonized the functional connectivity measures in their tangent space and then built brain age prediction models on the harmonized functional connectivity measures. We compared the brain age prediction models with alternatives that were built on the functional connectivity measures computed at a single scale and harmonized using different strategies. Comparison results have demonstrated that the best brain age prediction performance was achieved by the prediction model built on the multi-scale functional connectivity measures that were harmonized in tangent space, indicating that multi-scale functional connectivity measures provided richer information than those computed at any single scales and the harmonization of functional connectivity measures in tangent space improved the brain age prediction.

The accuracy of T1-weighted voxel-wise and region-wise metrics for brain age estimation

INTRODUCTION

The brain age score has recently been introduced for robust monitoring of brain morphological alterations throughout the lifespan, prediction of mortality risk, and early detection of neurological disorders.

METHODS

We assessed the brain age prediction accuracy of the widely used T1-weighted voxel-wise and region-wise metrics (i.e., T1-weighted magnetic resonance imaging [MRI]-wise metrics)) separately and their integration. We assessed 788 healthy individuals (age, 18-94 years) in a training set to build a brain age estimation framework based on different T1-weighted MRI-wise metrics (15 different metrics in total) and then validated each T1-weighted MRI-wise metric in an independent test set comprising 88 healthy individuals. We also assessed the accuracy of each T1-weighted MRI-wise metric in a clinical set of 70 patients with mild cognitive impairment and another of 30 patients with Alzheimer's disease.

RESULTS

Integration of gray matter voxel-wise maps and all region-wise metrics achieved the highest brain age prediction accuracy (mean absolute error, 4.63 years). These metrics on their own achieved lower accuracy (mean absolute error, 4.97 years and 5.75 years, respectively).

DISCUSSION

For tracing brain atrophy levels in neurological disorders at the clinical level, integration of voxel-wise and region-wise metrics may contribute to a more sensitive brain age framework than when these metrics are used on their own.

Brain-Predicted Age Difference Moderates the Association Between Muscle Strength and Mobility

BACKGROUND

Approximately 35% of individuals over age 70 report difficulty with mobility. Muscle weakness has been demonstrated to be one contributor to mobility limitations in older adults. The purpose of this study was to examine the moderating effect of brain-predicted age difference (an index of biological brain age/health derived from structural neuroimaging) on the relationship between leg strength and mobility.

METHODS

In community dwelling older adults (N = 57, 74.7 ± 6.93 years; 68% women), we assessed the relationship between isokinetic leg extensor strength and a composite measure of mobility [mobility battery assessment (MBA)] using partial Pearson correlations and multifactorial regression modeling. Brain predicted age (BPA) was calculated from T1 MR-images using a validated machine learning Gaussian Process regression model to explore the moderating effect of BPA difference (BPAD; BPA minus chronological age).

RESULTS

Leg strength was significantly correlated with BPAD (r = -0.317, p < 0.05) and MBA score (r = 0.541, p < 0.001). Chronological age, sex, leg strength, and BPAD explained 63% of the variance in MBA performance (p < 0.001). BPAD was a significant moderator of the relationship between strength and MBA, accounting for 7.0% of MBA score variance [△R 2 = 0.044, F(1,51) = 6.83, p = 0.01]. Conditional moderation effects of BPAD indicate strength was a stronger predictor of mobility in those with a great BPAD.

CONCLUSION

The relationship between strength and mobility appears to be influenced by brain aging, with strength serving as a possible compensation for decline in neural integrity.

Adipose tissue distribution from body MRI is associated with cross-sectional and longitudinal brain age in adults

There is an intimate body-brain connection in ageing, and obesity is a key risk factor for poor cardiometabolic health and neurodegenerative conditions. Although research has demonstrated deleterious effects of obesity on brain structure and function, the majority of studies have used conventional measures such as waist-to-hip ratio, waist circumference, and body mass index. While sensitive to gross features of body composition, such global anthropometric features fail to describe regional differences in body fat distribution and composition. The sample consisted of baseline brain magnetic resonance imaging (MRI) acquired from 790 healthy participants aged 18-94 years (mean ± standard deviation (SD) at baseline: 46.8 ± 16.3), and follow-up brain MRI collected from 272 of those individuals (two time-points with 19.7 months interval, on average (min = 9.8, max = 35.6). Of the 790 included participants, cross-sectional body MRI data was available from a subgroup of 286 participants, with age range 19-86 (mean = 57.6, SD = 15.6). Adopting a mixed cross-sectional and longitudinal design, we investigated cross-sectional body magnetic resonance imaging measures of adipose tissue distribution in relation to longitudinal brain structure using MRI-based morphometry (T1) and diffusion tensor imaging (DTI). We estimated tissue-specific brain age at two time points and performed Bayesian multilevel modelling to investigate the associations between adipose measures at follow-up and brain age gap (BAG) - the difference between actual age and the prediction of the brain's biological age - at baseline and follow-up. We also tested for interactions between BAG and both time and age on each adipose measure. The results showed credible associations between T1-based BAG and liver fat, muscle fat infiltration (MFI), and weight-to-muscle ratio (WMR), indicating older-appearing brains in people with higher measures of adipose tissue. Longitudinal evidence supported interaction effects between time and MFI and WMR on T1-based BAG, indicating accelerated ageing over the course of the study period in people with higher measures of adipose tissue. The results show that specific measures of fat distribution are associated with brain ageing and that different compartments of adipose tissue may be differentially linked with increased brain ageing, with potential to identify key processes involved in age-related transdiagnostic disease processes.

A catalogue of omics biological ageing clocks reveals substantial commonality and associations with disease risk

Biological age (BA), a measure of functional capacity and prognostic of health outcomes that discriminates between individuals of the same chronological age (chronAge), has been estimated using a variety of biomarkers. Previous comparative studies have mainly used epigenetic models (clocks), we use ~1000 participants to compare fifteen omics ageing clocks, with correlations of 0.21-0.97 with chronAge, even with substantial sub-setting of biomarkers. These clocks track common aspects of ageing with 95% of the variance in chronAge being shared among clocks. The difference between BA and chronAge - omics clock age acceleration (OCAA) - often associates with health measures. One year’s OCAA typically has the same effect on risk factors/10-year disease incidence as 0.09/0.25 years of chronAge. Epigenetic and IgG glycomics clocks appeared to track generalised ageing while others capture specific risks. We conclude BA is measurable and prognostic and that future work should prioritise health outcomes over chronAge.

Deep Learning in Large and Multi-Site Structural Brain MR Imaging Datasets

Large, multi-site, heterogeneous brain imaging datasets are increasingly required for the training, validation, and testing of advanced deep learning (DL)-based automated tools, including structural magnetic resonance (MR) image-based diagnostic and treatment monitoring approaches. When assembling a number of smaller datasets to form a larger dataset, understanding the underlying variability between different acquisition and processing protocols across the aggregated dataset (termed “batch effects”) is critical. The presence of variation in the training dataset is important as it more closely reflects the true underlying data distribution and, thus, may enhance the overall generalizability of the tool. However, the impact of batch effects must be carefully evaluated in order to avoid undesirable effects that, for example, may reduce performance measures. Batch effects can result from many sources, including differences in acquisition equipment, imaging technique and parameters, as well as applied processing methodologies. Their impact, both beneficial and adversarial, must be considered when developing tools to ensure that their outputs are related to the proposed clinical or research question (i.e., actual disease-related or pathological changes) and are not simply due to the peculiarities of underlying batch effects in the aggregated dataset. We reviewed applications of DL in structural brain MR imaging that aggregated images from neuroimaging datasets, typically acquired at multiple sites. We examined datasets containing both healthy control participants and patients that were acquired using varying acquisition protocols. First, we discussed issues around Data Access and enumerated the key characteristics of some commonly used publicly available brain datasets. Then we reviewed methods for correcting batch effects by exploring the two main classes of approaches: Data Harmonization that uses data standardization, quality control protocols or other similar algorithms and procedures to explicitly understand and minimize unwanted batch effects; and Domain Adaptation that develops DL tools that implicitly handle the batch effects by using approaches to achieve reliable and robust results. In this narrative review, we highlighted the advantages and disadvantages of both classes of DL approaches, and described key challenges to be addressed in future studies.

A comparison between diffusion tensor imaging and generalized q-sampling imaging in the age prediction of healthy adults via machine learning approaches

OBJECTIVE

Brain age, which is predicted using neuroimaging data, has become an important biomarker in aging research. This study applied diffusion tensor imaging (DTI) and generalized q-sampling imaging (GQI) model to predict age respectively, with the purpose of evaluating which diffusion model is more accurate in estimating age and revealing age-related changes in the brain.

APPROACH

Diffusion MRI data of 125 subjects from two sites were collected. Fractional anisotropy (FA) and quantitative anisotropy (QA) from the two diffusion models were calculated and were used as features of machine learning models. Sequential backward elimination algorithm was used for feature selection. Six machine learning approaches including linear regression, ridge regression, support vector regression (SVR) with linear kernel, quadratic kernel and radial basis function (RBF) kernel and feedforward neural network were used to predict age using FA and QA features respectively.

MAIN RESULTS

Age predictions using FA features were more accurate than predictions using QA features for all the 6 machine learning algorithms. Post-hoc analysis revealed that FA was more sensitive to age-related white matter alterations in the brain. In addition, SVR with RBF kernel based on FA features achieved better performances than the competing algorithms with MAE ranging from 7.74 to 10.54, MSE ranging from 87.79 to 150.86, and nMSE ranging from 0.05 to 0.14 Significance: FA from DTI model was more suitable than QA from GQI model in age prediction. FA metric was more sensitive to age-related white matter changes in the brain and FA of several brain regions could be used as white matter biomarkers in aging.

An uncertainty-aware, shareable, and transparent neural network architecture for brain-age modeling

The deviation between chronological age and age predicted from neuroimaging data has been identified as a sensitive risk marker of cross-disorder brain changes, growing into a cornerstone of biological age research. However, machine learning models underlying the field do not consider uncertainty, thereby confounding results with training data density and variability. Also, existing models are commonly based on homogeneous training sets, often not independently validated, and cannot be shared because of data protection issues. Here, we introduce an uncertainty-aware, shareable, and transparent Monte Carlo dropout composite quantile regression (MCCQR) Neural Network trained on N = 10,691 datasets from the German National Cohort. The MCCQR model provides robust, distribution-free uncertainty quantification in high-dimensional neuroimaging data, achieving lower error rates compared with existing models. In two examples, we demonstrate that it prevents spurious associations and increases power to detect deviant brain aging. We make the pretrained model and code publicly available.

Prediction and Modeling of Neuropsychological Scores in Alzheimer's Disease Using Multimodal Neuroimaging Data and Artificial Neural Networks

Background: In recent years, predicting and modeling the progression of Alzheimer's disease (AD) based on neuropsychological tests has become increasingly appealing in AD research. Objective: In this study, we aimed to predict the neuropsychological scores and investigate the non-linear progression trend of the cognitive declines based on multimodal neuroimaging data. Methods: We utilized unimodal/bimodal neuroimaging measures and a non-linear regression method (based on artificial neural networks) to predict the neuropsychological scores in a large number of subjects (n = 1143), including healthy controls (HC) and patients with mild cognitive impairment non-converter (MCI-NC), mild cognitive impairment converter (MCI-C), and AD. We predicted two neuropsychological scores, i.e., the clinical dementia rating sum of boxes (CDRSB) and Alzheimer's disease assessment scale cognitive 13 (ADAS13), based on structural magnetic resonance imaging (sMRI) and positron emission tomography (PET) biomarkers. Results: Our results revealed that volumes of the entorhinal cortex and hippocampus and the average fluorodeoxyglucose (FDG)-PET of the angular gyrus, temporal gyrus, and posterior cingulate outperform other neuroimaging features in predicting ADAS13 and CDRSB scores. Compared to a unimodal approach, our results showed that a bimodal approach of integrating the top two neuroimaging features (i.e., the entorhinal volume and the average FDG of the angular gyrus, temporal gyrus, and posterior cingulate) increased the prediction performance of ADAS13 and CDRSB scores in the converting and stable stages of MCI and AD. Finally, a non-linear AD progression trend was modeled to describe the cognitive decline based on neuroimaging biomarkers in different stages of AD. Conclusion: Findings in this study show an association between neuropsychological scores and sMRI and FDG-PET biomarkers from normal aging to severe AD.

Worse Physical Disability Is Associated With the Expression of PD-1 on Inflammatory T-Cells in Multiple Sclerosis Patients With Older Appearing Brains

Background: Magnetic Resonance Imaging (MRI) analysis method "brain-age" paradigm could offer an intuitive prognostic metric (brain-predicted age difference: brain-PAD) for disability in Multiple Sclerosis (MS), reflecting structural brain health adjusted for aging. Equally, cellular senescence has been reported in MS using T-cell biomarker CD8+CD57+. Objective: Here we explored links between MRI-derived brain-age and blood-derived cellular senescence. We examined the value of combining brain-PAD with CD8+CD57+(ILT2+PD-1+) T-cells when predicting disability score in MS and considered whether age-related biological mechanisms drive disability. Methods: Brain-age analysis was applied to T1-weighted MRI images. Disability was assessed and peripheral blood was examined for CD8+CD57+ T-cell phenotypes. Linear regression models were used, adjusted for sex, age and normalized brain volume. Results: We included 179 mainly relapsing-remitting MS patients. A high brain-PAD was associated with high physical disability (mean brain-PAD = +6.54 [5.12-7.95]). CD8+CD57+(ILT2+PD-1+) T-cell frequency was neither associated with disability nor with brain-PAD. Physical disability was predicted by the interaction between brain-PAD and CD8+CD57+ILT2+PD-1+ T-cell frequency (AR 2 = 0.196), yet without improvement compared to brain-PAD alone (AR 2 = 0.206; AICc = 1.8). Conclusion: Higher frequency of CD8+CD57+ILT2+PD-1+ T-cells in the peripheral blood in patients with an older appearing brain was associated with worse disability scores, suggesting a role of these cells in the development of disability in MS patients with poorer brain health.

Influence of endogenous estrogen on a network model of female brain integrity

Recent reports document sex differences in midlife brain integrity and metabolic health, such that more relationships are detectable between metabolic syndrome (MetS) components and markers of brain health in females than in males. Midlife is characterized by a rapid decrease in endogenous estrogen levels for women which is thought to increase risk for cardiometabolic disease and neurocognitive decline. Our study used network models, designed to explore the interconnectedness and organization of relationships among many variables at once, to compare the influence of endogenous estrogen and chronological age on a network of brain and metabolic health in order to investigate the utility of estrogen as a biomarker for brain vulnerability. Data were analyzed from 82 females (ages 40-62). Networks consisted of known biomarkers of risk for late-life cognitive decline: the five components of MetS; Brain-predicted age difference calculated on gray and white matter volume; white matter hyperintensities; Default Mode Network functional connectivity; cerebral concentrations of N-acetyl aspartate, glutamate and myo-inositol; and serum concentrations of estradiol. A second network replaced estradiol with chronological age. Expected influence (EI) of estradiol on the network was -1.190, relative to chronological age at -0.524, indicating that estradiol had a stronger expected influence over the network than age. A negative expected influence indicates that higher levels of estradiol would be expected to decrease the number of relationships in the model, which is thought to indicate lower risk. Overall, levels of estradiol appear more influential than chronological age at midlife for relationships between brain integrity and metabolic health.

Accelerated Brain Aging in Patients With Obsessive-Compulsive Disorder

Obsessive-compulsive disorder (OCD) may be accompanied by an accelerated structural decline of the brain with age compared to healthy controls (HCs); however, this has yet to be proven. To answer this question, we built a brain age prediction model using mean gray matter volumes of each brain region as features, which were obtained by voxel-based morphometry derived from T1-weighted MRI scans. The prediction model was built using two Chinese Han datasets (dataset 1, N = 106 for HCs and N = 90 for patients with OCD; dataset 2, N = 270 for HCs) to evaluate its performance. Then, a new prediction model was trained using data for HCs in dataset 1 and applied to patients with OCD to investigate the brain aging trajectory. The brain-predicted age difference (brain-PAD) scores, defined as the difference between predicted brain age and chronological age, were calculated for all participants and compared between patients with matched HCs in dataset 1. It was demonstrated that the prediction model performs consistently across different datasets. Patients with OCD presented higher brain-PAD scores than matched HCs, suggesting that patients with OCD presented accelerated brain aging. In addition, brain-PAD scores were negatively correlated with the duration of illness, suggesting that brain-PAD scores might capture progressive structural brain changes. These results identified accelerated brain aging in patients with OCD for the first time and deepened our understanding of the pathogenesis of OCD.

Associations between alcohol use and accelerated biological ageing

Harmful alcohol use is a leading cause of premature death and is associated with age-related disease. Biological ageing is highly variable between individuals and may deviate from chronological ageing, suggesting that biomarkers of biological ageing (derived from DNA methylation or brain structural measures) may be clinically relevant. Here, we investigated the relationships between alcohol phenotypes and both brain and DNA methylation age estimates. First, using data from UK Biobank and Generation Scotland, we tested the association between alcohol consumption (units/week) or hazardous use (Alcohol Use Disorders Identification Test [AUDIT] scores) and accelerated brain and epigenetic ageing in 20,258 and 8051 individuals, respectively. Second, we used Mendelian randomisation (MR) to test for a causal effect of alcohol consumption levels and alcohol use disorder (AUD) on biological ageing. Alcohol use showed a consistent positive association with higher predicted brain age (AUDIT-C: β = 0.053, p = 3.16 × 10-13 ; AUDIT-P: β = 0.052, p = 1.6 × 10-13 ; total AUDIT score: β = 0.062, p = 5.52 × 10-16 ; units/week: β = 0.078, p = 2.20 × 10-16 ), and two DNA methylation-based estimates of ageing, GrimAge (units/week: β = 0.053, p = 1.48 × 10-7 ) and PhenoAge (units/week: β = 0.077, p = 2.18x10-10 ). MR analyses revealed limited evidence for a causal effect of AUD on accelerated brain ageing (β = 0.118, p = 0.044). However, this result should be interpreted cautiously as the significant effect was driven by a single genetic variant. We found no evidence for a causal effect of alcohol consumption levels on accelerated biological ageing. Future studies investigating the mechanisms associating alcohol use with accelerated biological ageing are warranted.

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.

NeuroCrypt: Machine Learning Over Encrypted Distributed Neuroimaging Data

The field of neuroimaging can greatly benefit from building machine learning models to detect and predict diseases, and discover novel biomarkers, but much of the data collected at various organizations and research centers is unable to be shared due to privacy or regulatory concerns (especially for clinical data or rare disorders). In addition, aggregating data across multiple large studies results in a huge amount of duplicated technical debt and the resources required can be challenging or impossible for an individual site to build. Training on the data distributed across organizations can result in models that generalize much better than models trained on data from any of organizations alone. While there are approaches for decentralized sharing, these often do not provide the highest possible guarantees of sample privacy that only cryptography can provide. In addition, such approaches are often focused on probabilistic solutions. In this paper, we propose an approach that leverages the potential of datasets spread among a number of data collecting organizations by performing joint analyses in a secure and deterministic manner when only encrypted data is shared and manipulated. The approach is based on secure multiparty computation which refers to cryptographic protocols that enable distributed computation of a function over distributed inputs without revealing additional information about the inputs. It enables multiple organizations to train machine learning models on their joint data and apply the trained models to encrypted data without revealing their sensitive data to the other parties. In our proposed approach, organizations (or sites) securely collaborate to build a machine learning model as it would have been trained on the aggregated data of all the organizations combined. Importantly, the approach does not require a trusted party (i.e. aggregator), each contributing site plays an equal role in the process, and no site can learn individual data of any other site. We demonstrate effectiveness of the proposed approach, in a range of empirical evaluations using different machine learning algorithms including logistic regression and convolutional neural network models on human structural and functional magnetic resonance imaging datasets.