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 predicted using graph convolutional neural network explains neurodevelopmental trajectory in preterm neonates

OBJECTIVES

Dramatic brain morphological changes occur throughout the third trimester of gestation. In this study, we investigated whether the predicted brain age (PBA) derived from graph convolutional network (GCN) that accounts for cortical morphometrics in third trimester is associated with postnatal abnormalities and neurodevelopmental outcome.

METHODS

In total, 577 T1 MRI scans of preterm neonates from two different datasets were analyzed; the NEOCIVET pipeline generated cortical surfaces and morphological features, which were then fed to the GCN to predict brain age. The brain age index (BAI; PBA minus chronological age) was used to determine the relationships among preterm birth (i.e., birthweight and birth age), perinatal brain injuries, postnatal events/clinical conditions, BAI at postnatal scan, and neurodevelopmental scores at 30 months.

RESULTS

Brain morphology and GCN-based age prediction of preterm neonates without brain lesions (mean absolute error [MAE]: 0.96 weeks) outperformed conventional machine learning methods using no topological information. Structural equation models (SEM) showed that BAI mediated the influence of preterm birth and postnatal clinical factors, but not perinatal brain injuries, on neurodevelopmental outcome at 30 months of age.

CONCLUSIONS

Brain morphology may be clinically meaningful in measuring brain age, as it relates to postnatal factors, and predicting neurodevelopmental outcome.

CLINICAL RELEVANCE STATEMENT

Understanding the neurodevelopmental trajectory of preterm neonates through the prediction of brain age using a graph convolutional neural network may allow for earlier detection of potential developmental abnormalities and improved interventions, consequently enhancing the prognosis and quality of life in this vulnerable population.

KEY POINTS

•Brain age in preterm neonates predicted using a graph convolutional network with brain morphological changes mediates the pre-scan risk factors and post-scan neurodevelopmental outcomes. •Predicted brain age oriented from conventional deep learning approaches, which indicates the neurodevelopmental status in neonates, shows a lack of sensitivity to perinatal risk factors and predicting neurodevelopmental outcomes. •The new brain age index based on brain morphology and graph convolutional network enhances the accuracy and clinical interpretation of predicted brain age for neonates.

Brain Age Prediction in Developing Childhood with Multimodal Magnetic Resonance Images

It is well known that brain development is very fast and complex in the early childhood with age-based neurological and physiological changes of brain structure and function. The brain maturity is an important indicator for evaluating the normal development of children. In this paper, we propose a multimodal regression framework to combine the features from structural magnetic resonance imaging (sMRI) and diffusion tensor imaging (DTI) data for age prediction of children. First, three types of features are extracted from sMRI and DTI data. Second, we propose to combine the sparse coding and Q-Learning for feature selection from each modality. Finally, the ensemble regression is performed by random forest based on proximity measures to fuse multimodal features for age prediction. The proposed method is evaluated on 212 participants, including 76 young children less than 2 years old and 136 children aged from 2-15 years old recruited from Shanghai Children's Hospital. The results show that integrating multimodal features has achieved the highest accuracies with the root mean squared error (RMSE) of 0.208 years and mean absolute error (MAE) of 0.150 years for age prediction of young children (0-2), and RMSE of 1.666 years and MAE of 1.087 years for older children (2-15). We have shown that the selected features by Q-Learning can consistently improve the prediction accuracy. The comparison of prediction results demonstrates that the proposed method performs better than other competing methods.

Dose-dependent relationship between social drinking and brain aging

Low-level alcohol consumption is commonly perceived as being inconsequential or even beneficial for overall health, with some reports suggesting that it may protect against dementia or cardiovascular risks. However, these potential benefits do not preclude the concurrent possibility of negative health outcomes related to alcohol consumption. To examine whether casual, non-heavy drinking is associated with premature brain aging, we utilized the Brain-Age Regression Analysis and Computational Utility Software package to predict brain age in a community sample of adults [n = 240, mean age 35.1 (±10.7) years, 48% male, 49% African American]. Accelerated brain aging was operationalized as the difference between predicted and chronological age ("brain age gap"). Multiple regression analysis revealed a significant association between previous 90-day alcohol consumption and brain age gap (β = 0.014, p = 0.023). We replicated these results in an independent cohort [n = 231 adults, mean age 34.3 (±11.1) years, 55% male, 28% African American: β = 0.014, p = 0.002]. Our results suggest that even low-level alcohol consumption is associated with premature brain aging. The clinical significance of these findings remains to be investigated.

Long-term Anabolic-Androgenic Steroid Use Is Associated With Deviant Brain Aging

BACKGROUND

High-dose long-term use of anabolic-androgenic steroids (AASs) may cause a range of adverse effects, including brain and cognitive abnormalities. We performed age prediction based on brain scans to test whether prolonged AAS use is associated with accentuated brain aging.

METHODS

T1-weighted magnetic resonance imaging (3D MPRAGE [magnetization-prepared rapid acquisition gradient-echo]) scans were obtained from male weightlifters with a history of prolonged AAS use (n = 130) or no AAS use (n = 99). We trained machine learning models on combinations of regional brain volumes, cortical thickness, and surface area in an independent training set of 1838 healthy male subjects (18-92 years of age) and predicted brain age for each participant in our study. Including cross-sectional and longitudinal (mean interval = 3.5 years, n = 76) magnetic resonance imaging data, we used linear mixed-effects models to compare the gap between chronological age and predicted brain age (the brain age gap [BAG]) for the two groups and tested for group differences in the rate of change in BAG. We tested for associations between apparent brain aging and AAS use duration, pattern of administration, and dependence.

RESULTS

AAS users had higher BAG compared with weightlifting control subjects, which was associated with dependency and longer history of use. Group differences in BAG could not be explained by other substance use, general cognitive abilities, or depression. While longitudinal analysis revealed no evidence of increased brain aging in the overall AAS group, accelerated brain aging was seen with longer AAS exposure.

CONCLUSIONS

The findings suggest that long-term high-dose AAS use may have adverse effects on brain aging, potentially linked to dependency and exaggerated use of AASs.

Brain age prediction in schizophrenia: Does the choice of machine learning algorithm matter?

Brain-predicted age difference (brainPAD) has been used in schizophrenia to assess individual-level deviation in the biological age of the patients' brain (i.e., brain-age) from normative reference brain structural datasets. There is marked inter-study variation in brainPAD in schizophrenia which is commonly attributed to sample heterogeneity. However, the potential contribution of the different machine learning algorithms used for brain-age estimation has not been systematically evaluated. Here, we aimed to assess variation in brain-age estimated by six commonly used algorithms [ordinary least squares regression, ridge regression, least absolute shrinkage and selection operator regression, elastic-net regression, linear support vector regression, and relevance vector regression] when applied to the same brain structural features from the same sample. To assess reproducibility we used data from two publically available samples of healthy individuals (n = 1092 and n = 492) and two further samples, from the Icahn School of Medicine at Mount Sinai (ISMMS) and the Center of Biomedical Research Excellence (COBRE), comprising both patients with schizophrenia (n = 90 and n = 76) and healthy individuals (n = 200 and n = 87). Performance similarity across algorithms was compared within each sample using correlation analyses and hierarchical clustering. Across all samples ordinary least squares regression, the only algorithm without a penalty term, performed markedly worse. All other algorithms showed comparable performance but they still yielded variable brain-age estimates despite being applied to the same data. Although brainPAD was consistently higher in patients with schizophrenia, it varied by algorithm from 3.8 to 5.2 years in the ISMMS sample and from to 4.5 to 11.7 years in the COBRE sample. Algorithm choice introduces variations in brain-age and may confound inter-study comparisons when assessing brainPAD in schizophrenia.

Prediction of brain age from routine T2-weighted spin-echo brain magnetic resonance images with a deep convolutional neural network

Our study investigated the feasibility and clinical relevance of brain age prediction using axial T2-weighted images (T2-WIs) with a deep convolutional neural network (CNN) algorithm. The CNN model was trained by 1,530 scans in our institution. The performance was evaluated by the mean absolute error (MAE) between the predicted brain age and the chronological age based on an internal test set (n=270) and an external test set (n=560). The ensemble CNN model showed an MAE of 4.22 years in the internal test set and 9.96 years in the external test set. Participants with grade 2-3 white matter hyperintensity (WMH) showed a higher corrected predicted age difference (PAD) than grade 0 WMH (posthoc p<0.001). Participants diagnosed with diabetes mellitus also had a higher corrected PAD than those without diabetes (adjusted p=0.048), although it showed no significant differences according to the diagnosis of hypertension or dyslipidemia. We suggest that routine clinical T2-WIs are feasible to predict brain age, and it might be clinically relevant according to the WMH grade and the presence of diabetes mellitus.

Accurate brain age prediction with lightweight deep neural networks

Deep learning has huge potential for accurate disease prediction with neuroimaging data, but the prediction performance is often limited by training-dataset size and computing memory requirements. To address this, we propose a deep convolutional neural network model, Simple Fully Convolutional Network (SFCN), for accurate prediction of brain age using T1-weighted structural MRI data. Compared with other popular deep network architectures, SFCN has fewer parameters, so is more compatible with small dataset size and 3D volume data. The network architecture was combined with several techniques for boosting performance, including data augmentation, pre-training, model regularization, model ensemble and prediction bias correction. We compared our overall SFCN approach with several widely-used machine learning models. It achieved state-of-the-art performance in UK Biobank data (N = 14,503), with mean absolute error (MAE) = 2.14y in brain age prediction and 99.5% in sex classification. SFCN also won (both parts of) the 2019 Predictive Analysis Challenge for brain age prediction, involving 79 competing teams (N = 2,638, MAE = 2.90y). We describe here the details of our approach, and its optimisation and validation. Our approach can easily be generalised to other tasks using different image modalities, and is released on GitHub.

Multimodal brain-age prediction and cardiovascular risk: The Whitehall II MRI sub-study

Brain age is becoming a widely applied imaging-based biomarker of neural aging and potential proxy for brain integrity and health. We estimated multimodal and modality-specific brain age in the Whitehall II (WHII) MRI cohort using machine learning and imaging-derived measures of gray matter (GM) morphology, white matter microstructure (WM), and resting state functional connectivity (FC). The results showed that the prediction accuracy improved when multiple imaging modalities were included in the model (R2 = 0.30, 95% CI [0.24, 0.36]). The modality-specific GM and WM models showed similar performance (R2 = 0.22 [0.16, 0.27] and R2 = 0.24 [0.18, 0.30], respectively), while the FC model showed the lowest prediction accuracy (R2 = 0.002 [-0.005, 0.008]), indicating that the FC features were less related to chronological age compared to structural measures. Follow-up analyses showed that FC predictions were similarly low in a matched sub-sample from UK Biobank, and although FC predictions were consistently lower than GM predictions, the accuracy improved with increasing sample size and age range. Cardiovascular risk factors, including high blood pressure, alcohol intake, and stroke risk score, were each associated with brain aging in the WHII cohort. Blood pressure showed a stronger association with white matter compared to gray matter, while no differences in the associations of alcohol intake and stroke risk with these modalities were observed. In conclusion, machine-learning based brain age prediction can reduce the dimensionality of neuroimaging data to provide meaningful biomarkers of individual brain aging. However, model performance depends on study-specific characteristics including sample size and age range, which may cause discrepancies in findings across studies.

Brain status modeling with non-negative projective dictionary learning

Accurate prediction of individuals' brain age is critical to establish a baseline for normal brain development. This study proposes to model brain development with a novel non-negative projective dictionary learning (NPDL) approach, which learns a discriminative representation of multi-modal neuroimaging data for predicting brain age. Our approach encodes the variability of subjects in different age groups using separate dictionaries, projecting features into a low-dimensional manifold such that information is preserved only for the corresponding age group. The proposed framework improves upon previous discriminative dictionary learning methods by incorporating orthogonality and non-negativity constraints, which remove representation redundancy and perform implicit feature selection. We study brain development on multi-modal brain imaging data from the PING dataset (N = 841, age = 3-21 years). The proposed analysis uses our NDPL framework to predict the age of subjects based on cortical measures from T1-weighted MRI and connectome from diffusion weighted imaging (DWI). We also investigate the association between age prediction and cognition, and study the influence of gender on prediction accuracy. Experimental results demonstrate the usefulness of NDPL for modeling brain development.

Brain age prediction in stroke patients: Highly reliable but limited sensitivity to cognitive performance and response to cognitive training

Cognitive deficits are important predictors for outcome, independence and quality of life after stroke, but often remain unnoticed and unattended because other impairments are more evident. Computerized cognitive training (CCT) is among the candidate interventions that may alleviate cognitive difficulties, but the evidence supporting its feasibility and effectiveness is scarce, partly due to the lack of tools for outcome prediction and monitoring. Magnetic resonance imaging (MRI) provides candidate markers for disease monitoring and outcome prediction. By integrating information not only about lesion extent and localization, but also regarding the integrity of the unaffected parts of the brain, advanced MRI provides relevant information for developing better prediction models in order to tailor cognitive intervention for patients, especially in a chronic phase. Using brain age prediction based on MRI based brain morphometry and machine learning, we tested the hypotheses that stroke patients with a younger-appearing brain relative to their chronological age perform better on cognitive tests and benefit more from cognitive training compared to patients with an older-appearing brain. In this randomized double-blind study, 54 patients who suffered mild stroke (>6 months since hospital admission, NIHSS≤7 at hospital discharge) underwent 3-weeks CCT and MRI before and after the intervention. In addition, patients were randomized to one of two groups receiving either active or sham transcranial direct current stimulation (tDCS). We tested for main effects of brain age gap (estimated age - chronological age) on cognitive performance, and associations between brain age gap and task improvement. Finally, we tested if longitudinal changes in brain age gap during the intervention were sensitive to treatment response. Briefly, our results suggest that longitudinal brain age prediction based on automated brain morphometry is feasible and reliable in stroke patients. However, no significant association between brain age and both performance and response to cognitive training were found.

Brain age prediction: Cortical and subcortical shape covariation in the developing human brain

Cortical development is characterized by distinct spatial and temporal patterns of maturational changes across various cortical shape measures. There is a growing interest in summarizing complex developmental patterns into a single index, which can be used to characterize an individual's brain age. We conducted this study with two primary aims. First, we sought to quantify covariation patterns for a variety of cortical shape measures, including cortical thickness, gray matter volume, surface area, mean curvature, and travel depth, as well as white matter volume, and subcortical gray matter volume. We examined these measures in a sample of 869 participants aged 5-18 from the Healthy Brain Network (HBN) neurodevelopmental cohort using the Joint and Individual Variation Explained (Lock et al., 2013) method. We validated our results in an independent dataset from the Nathan Kline Institute - Rockland Sample (NKI-RS; N = 210) and found remarkable consistency for some covariation patterns. Second, we assessed whether covariation patterns in the brain can be used to accurately predict a person's chronological age. Using ridge regression, we showed that covariation patterns can predict chronological age with high accuracy, reflected by our ability to cross-validate our model in an independent sample with a correlation coefficient of 0.84 between chronologic and predicted age. These covariation patterns also predicted sex with high accuracy (AUC = 0.85), and explained a substantial portion of variation in full scale intelligence quotient (R² = 0.10). In summary, we found significant covariation across different cortical shape measures and subcortical gray matter volumes. In addition, each shape measure exhibited distinct covariations that could not be accounted for by other shape measures. These covariation patterns accurately predicted chronological age, sex and general cognitive ability. In a subset of NKI-RS, test-retest (<1 month apart, N = 120) and longitudinal scans (1.22 ± 0.29 years apart, N = 77) were available, allowing us to demonstrate high reliability for the prediction models obtained and the ability to detect subtle differences in the longitudinal scan interval among participants (median and median absolute deviation of absolute differences between predicted age difference and real age difference = 0.53 ± 0.47 years, r = 0.24, p-value = 0.04).