A deep learning model for brain age prediction using minimally preprocessed T1w images as input

Decomposing the effect of normal aging and Alzheimer's disease in brain morphological changes via learned aging templates

Alzheimer's disease (AD) subjects usually show more profound morphological changes with time compared to cognitively normal (CN) individuals. These changes are the combination of two major biological processes: normal aging and AD pathology. Investigating normal aging and residual morphological changes separately can increase our understanding of the disease. This paper proposes two scores, the aging score (AS) and the AD-specific score (ADS), whose purpose is to measure these two components of brain atrophy independently. For this, in the first step, we estimate the atrophy due to the normal aging of CN subjects by computing the expected deformation required to match imaging templates generated at different ages. We used a state-of-the-art generative deep learning model for generating such imaging templates. In the second step, we apply deep learning-based diffeomorphic registration to align the given image of a subject with a reference imaging template. Parametrization of this deformation field is then decomposed voxel-wise into their parallel and perpendicular components with respect to the parametrization of the expected atrophy of CN individuals in one year computed in the first step. AS and ADS are the normalized scores of these two components, respectively. We evaluated these two scores on the OASIS-3 dataset with 1,014 T1-weighted MRI scans. Of these, 326 scans were from CN subjects, and 688 scans were from subjects diagnosed with AD at various stages of clinical severity, as defined by clinical dementia rating (CDR) scores. Our results reveal that AD is marked by both disease-specific brain changes and an accelerated aging process. Such changes affect brain regions differently. Moreover, the proposed scores were sensitive to detect changes in the early stages of the disease, which is promising for its potential future use in clinical studies. Our code is freely available at https://github.com/Fjr9516/DBM_with_DL .

UK Biobank MRI data can power the development of generalizable brain clocks: A study of standard ML/DL methodologies and performance analysis on external databases

In this study, we present a comprehensive pipeline to train and compare a broad spectrum of machine learning and deep learning brain clocks, integrating diverse preprocessing strategies and correction terms. Our analysis also includes established methodologies which have shown success in prior UK Biobank-related studies. For our analysis we used T1-weighted MRI scans and processed de novo all images via FastSurfer, transforming them into a conformed space for deep learning and extracting image-derived phenotypes for our machine learning approaches. We rigorously evaluated these approaches both as robust age predictors for healthy individuals and as potential biomarkers for various neurodegenerative conditions, leveraging data from the UK Biobank, ADNI, and NACC datasets. To this end we designed a statistical framework to assess age prediction performance, the robustness of the prediction across cohort variability (database, machine type and ethnicity) and its potential as a biomarker for neurodegenerative conditions. Results demonstrate that highly accurate brain age models, typically utilising penalised linear machine learning models adjusted with Zhang's methodology, with mean absolute errors under 1 year in external validation, can be achieved while maintaining consistent prediction performance across different age brackets and subgroups (e.g., ethnicity and MRI machine/manufacturer). Additionally, these models show strong potential as biomarkers for neurodegenerative conditions, such as dementia, where brain age prediction achieved an AUROC of up to 0.90 in distinguishing healthy individuals from those with dementia.

Biological brain age and resilience in cognitively unimpaired 70-year-old individuals

INTRODUCTION

This study investigated the associations of brain age gap (BAG)-a biological marker of brain resilience-with life exposures, neuroimaging measures, biological processes, and cognitive function.

METHODS

We derived BAG by subtracting predicted brain age from chronological age in 739 septuagenarians without dementia or neurological disorders. Robust linear regression models assessed BAG associations with life exposures, plasma inflammatory and metabolic biomarkers, magnetic resonance imaging, and cerebrospinal fluid biomarkers of neurodegeneration and vascular brain injury, and cognitive performance.

RESULTS

Greater BAG (older-looking brains) was associated with physical inactivity, diabetes, and stroke, while prediabetes was related to lower BAG, that is, younger-looking brains. Physical activity mitigated the link between obesity and BAG. Greater BAG was associated with greater small vessel disease burden, white-matter alterations, inflammation, high glucose, poorer vascular-related cognitive domains. Sex-specific associations were identified.

DISCUSSION

Vascular-related lifestyles and health shape brain appearance. Inflammation and insulin-related processes may be keys to understanding vascular cognitive disorders.

HIGHLIGHTS

BAG, reflecting deviations from CA, can indicate resilience. Diabetes, stroke, and low physical activity link to "older" brains (greater BAG). Physical activity yielded to "younger" brains in septuagenarians with obesity. High cerebrovascular burden, inflammation, and glucose associate with "older" brains. Sex differences were detected in all BAG-associated factors.

Multivariate Modelling and Prediction of High-Frequency Sensor-Based Cerebral Physiologic Signals: Narrative Review of Machine Learning Methodologies

Monitoring cerebral oxygenation and metabolism, using a combination of invasive and non-invasive sensors, is vital due to frequent disruptions in hemodynamic regulation across various diseases. These sensors generate continuous high-frequency data streams, including intracranial pressure (ICP) and cerebral perfusion pressure (CPP), providing real-time insights into cerebral function. Analyzing these signals is crucial for understanding complex brain processes, identifying subtle patterns, and detecting anomalies. Computational models play an essential role in linking sensor-derived signals to the underlying physiological state of the brain. Multivariate machine learning models have proven particularly effective in this domain, capturing intricate relationships among multiple variables simultaneously and enabling the accurate modeling of cerebral physiologic signals. These models facilitate the development of advanced diagnostic and prognostic tools, promote patient-specific interventions, and improve therapeutic outcomes. Additionally, machine learning models offer great flexibility, allowing different models to be combined synergistically to address complex challenges in sensor-based data analysis. Ensemble learning techniques, which aggregate predictions from diverse models, further enhance predictive accuracy and robustness. This review explores the use of multivariate machine learning models in cerebral physiology as a whole, with an emphasis on sensor-derived signals related to hemodynamics, cerebral oxygenation, metabolism, and other modalities such as electroencephalography (EEG) and functional near-infrared spectroscopy (fNIRS) where applicable. It will detail the operational principles, mathematical foundations, and clinical implications of these models, providing a deeper understanding of their significance in monitoring cerebral function.

Explainable brain age prediction: a comparative evaluation of morphometric and deep learning pipelines

Brain age, a biomarker reflecting brain health relative to chronological age, is increasingly used in neuroimaging to detect early signs of neurodegenerative diseases and support personalized treatment plans. Two primary approaches for brain age prediction have emerged: morphometric feature extraction from MRI scans and deep learning (DL) applied to raw MRI data. However, a systematic comparison of these methods regarding performance, interpretability, and clinical utility has been limited. In this study, we present a comparative evaluation of two pipelines: one using morphometric features from FreeSurfer and the other employing 3D convolutional neural networks (CNNs). Using a multisite neuroimaging dataset, we assessed both model performance and the interpretability of predictions through eXplainable Artificial Intelligence (XAI) methods, applying SHAP to the feature-based pipeline and Grad-CAM and DeepSHAP to the CNN-based pipeline. Our results show comparable performance between the two pipelines in Leave-One-Site-Out (LOSO) validation, achieving state-of-the-art performance on the independent test set ( M A E = 3.21 with DNN and morphometric features and M A E = 3.08 with a DenseNet-121 architecture). SHAP provided the most consistent and interpretable results, while DeepSHAP exhibited greater variability. Further work is needed to assess the clinical utility of Grad-CAM. This study addresses a critical gap by systematically comparing the interpretability of multiple XAI methods across distinct brain age prediction pipelines. Our findings underscore the importance of integrating XAI into clinical practice, offering insights into how XAI outputs vary and their potential utility for clinicians.

Evaluation of Brain Age as a Specific Marker of Brain Health

Brain age is a powerful marker of general brain health. Furthermore, brain age models are trained on large datasets, thus giving them a potential advantage in predicting specific outcomes - much like the success of finetuning large language models for specific applications. However, it is also well-accepted in machine learning that models trained to directly predict specific outcomes (i.e., direct models) often perform better than those trained on surrogate outcomes. Therefore, despite their much larger training data, it is unclear whether brain age models outperform direct models in predicting specific brain health outcomes. Here, we compare large-scale brain age models and direct models for predicting specific health outcomes in the context of Alzheimer's Disease (AD) dementia. Using anatomical T1 scans from three continents (N = 1,848), we find that direct models outperform brain age models without finetuning. Finetuned brain age models yielded similar performance as direct models, but importantly, did not outperform direct models although the brain age models were pretrained on 1000 times more data than the direct models: N = 53,542 vs N = 50. Overall, our results do not discount brain age as a useful marker of general brain health. However, in this era of large-scale brain age models, our results suggest that small-scale, targeted approaches for extracting specific brain health markers still hold significant value.

Anatomic Interpretability in Neuroimage Deep Learning: Saliency Approaches for Typical Aging and Traumatic Brain Injury

The black box nature of deep neural networks (DNNs) makes researchers and clinicians hesitant to rely on their findings. Saliency maps can enhance DNN explainability by suggesting the anatomic localization of relevant brain features. This study compares seven popular attribution-based saliency approaches to assign neuroanatomic interpretability to DNNs that estimate biological brain age (BA) from magnetic resonance imaging (MRI). Cognitively normal (CN) adults ( N = 13,394 , 5,900 males; mean age: 65.82 ± 8.89 years) are included for DNN training, testing, validation, and saliency map generation to estimate BA. To study saliency robustness to the presence of anatomic deviations from normality, saliency maps are also generated for adults with mild traumatic brain injury (mTBI, N = 214 , 135 males; mean age: 55.3 ± 9.9 years). We assess saliency methods' capacities to capture known anatomic features of brain aging and compare them to a surrogate ground truth whose anatomic saliency is known a priori. Anatomic aging features are identified most reliably by the integrated gradients method, which outperforms all others through its ability to localize relevant anatomic features. Gradient Shapley additive explanations, input × gradient, and masked gradient perform less consistently but still highlight ubiquitous neuroanatomic features of aging (ventricle dilation, hippocampal atrophy, sulcal widening). Saliency methods involving gradient saliency, guided backpropagation, and guided gradient-weight class attribution mapping localize saliency outside the brain, which is undesirable. Our research suggests the relative tradeoffs of saliency methods to interpret DNN findings during BA estimation in typical aging and after mTBI.

Brain age prediction via cross-stratified ensemble learning

As an important biomarker of neural aging, the brain age reflects the integrity and health of the human brain. Accurate prediction of brain age could help to understand the underlying mechanism of neural aging. In this study, a cross-stratified ensemble learning algorithm with staking strategy was proposed to obtain brain age and the derived predicted age difference (PAD) using T1-weighted magnetic resonance imaging (MRI) data. The approach was characterized as by implementing two modules: one was three base learners of 3D-DenseNet, 3D-ResNeXt, 3D-Inception-v4; another was 14 secondary learners of liner regressions. To evaluate performance, our method was compared with single base learners, regular ensemble learning algorithms, and state-of-the-art (SOTA) methods. The results demonstrated that our proposed model outperformed others models, with three metrics of mean absolute error (MAE), root mean-squared error (RMSE), and coefficient of determination (R2) of 2.9405 years, 3.9458 years, and 0.9597, respectively. Furthermore, there existed significant differences in PAD among the three groups of normal control (NC), mild cognitive impairment (MCI) and Alzheimer's disease (AD), with an increased trend across NC, MCI, and AD. It was concluded that the proposed algorithm could be effectively used in computing brain aging and PAD, and offering potential for early diagnosis and assessment of normal brain aging and AD.

Anatomic Interpretability in Neuroimage Deep Learning: Saliency Approaches for Typical Aging and Traumatic Brain Injury

The black box nature of deep neural networks (DNNs) makes researchers and clinicians hesitant to rely on their findings. Saliency maps can enhance DNN explainability by suggesting the anatomic localization of relevant brain features. This study compares seven popular attribution-based saliency approaches to assign neuroanatomic interpretability to DNNs that estimate biological brain age (BA) from magnetic resonance imaging (MRI). Cognitively normal (CN) adults (N = 13,394, 5,900 males; mean age: 65.82 ± 8.89 years) are included for DNN training, testing, validation, and saliency map generation to estimate BA. To study saliency robustness to the presence of anatomic deviations from normality, saliency maps are also generated for adults with mild traumatic brain injury (mTBI, N = 214, 135 males; mean age: 55.3 ± 9.9 years). We assess saliency methods' capacities to capture known anatomic features of brain aging and compare them to a surrogate ground truth whose anatomic saliency is known a priori. Anatomic aging features are identified most reliably by the integrated gradients method, which outperforms all others through its ability to localize relevant anatomic features. Gradient Shapley additive explanations, input × gradient, and masked gradient perform less consistently but still highlight ubiquitous neuroanatomic features of aging (ventricle dilation, hippocampal atrophy, sulcal widening). Saliency methods involving gradient saliency, guided backpropagation, and guided gradient-weight class attribution mapping localize saliency outside the brain, which is undesirable. Our research suggests the relative tradeoffs of saliency methods to interpret DNN findings during BA estimation in typical aging and after mTBI.

Analysis of Brain Age Gap across Subject Cohorts and Prediction Model Architectures

Background: Brain age prediction from brain MRI scans and the resulting brain age gap (BAG)-the difference between predicted brain age and chronological age-is a general biomarker for a variety of neurological, psychiatric, and other diseases or disorders. Methods: This study examined the differences in BAG values derived from T1-weighted scans using five state-of-the-art deep learning model architectures previously used in the brain age literature: 2D/3D VGG, RelationNet, ResNet, and SFCN. The models were evaluated on healthy controls and cohorts with sleep apnea, diabetes, multiple sclerosis, Parkinson's disease, mild cognitive impairment, and Alzheimer's disease, employing rigorous statistical analysis, including repeated model training and linear mixed-effects models. Results: All five models consistently identified a statistically significant positive BAG for diabetes (ranging from 0.79 years with RelationNet to 2.13 years with SFCN), multiple sclerosis (2.67 years with 3D VGG to 4.24 years with 2D VGG), mild cognitive impairment (2.13 years with 2D VGG to 2.59 years with 3D VGG), and Alzheimer's dementia (5.54 years with ResNet to 6.48 years with SFCN). For Parkinson's disease, a statistically significant BAG increase was observed in all models except ResNet (1.30 years with 2D VGG to 2.59 years with 3D VGG). For sleep apnea, a statistically significant BAG increase was only detected with the SFCN model (1.59 years). Additionally, we observed a trend of decreasing BAG with increasing chronological age, which was more pronounced in diseased cohorts, particularly those with the largest BAG, such as multiple sclerosis (-0.34 to -0.2), mild cognitive impairment (-0.37 to -0.26), and Alzheimer's dementia (-0.66 to -0.47), compared to healthy controls (-0.18 to -0.1). Conclusions: Consistent with previous research, Alzheimer's dementia and multiple sclerosis exhibited the largest BAG across all models, with SFCN predicting the highest BAG overall. The negative BAG trend suggests a complex interplay of survival bias, disease progression, adaptation, and therapy that influences brain age prediction across the age spectrum.

Assessing CT-based Volumetric Analysis via Transfer Learning with MRI and Manual Labels for Idiopathic Normal Pressure Hydrocephalus

Background

Brain computed tomography (CT) is an accessible and commonly utilized technique for assessing brain structure. In cases of idiopathic normal pressure hydrocephalus (iNPH), the presence of ventriculomegaly is often neuroradiologically evaluated by visual rating and manually measuring each image. Previously, we have developed and tested a deep-learning-model that utilizes transfer learning from magnetic resonance imaging (MRI) for CT-based intracranial tissue segmentation. Accordingly, herein we aimed to enhance the segmentation of ventricular cerebrospinal fluid (VCSF) in brain CT scans and assess the performance of automated brain CT volumetrics in iNPH patient diagnostics.

Methods

The development of the model used a two-stage approach. Initially, a 2D U-Net model was trained to predict VCSF segmentations from CT scans, using paired MR-VCSF labels from healthy controls. This model was subsequently refined by incorporating manually segmented lateral CT-VCSF labels from iNPH patients, building on the features learned from the initial U-Net model. The training dataset included 734 CT datasets from healthy controls paired with T1-weighted MRI scans from the Gothenburg H70 Birth Cohort Studies and 62 CT scans from iNPH patients at Uppsala University Hospital. To validate the model's performance across diverse patient populations, external clinical images including scans of 11 iNPH patients from the Universitatsmedizin Rostock, Germany, and 30 iNPH patients from the University of Alabama at Birmingham, United States were used. Further, we obtained three CT-based volumetric measures (CTVMs) related to iNPH.

Results

Our analyses demonstrated strong volumetric correlations (ϱ=0.91, p<0.001) between automatically and manually derived CT-VCSF measurements in iNPH patients. The CTVMs exhibited high accuracy in differentiating iNPH patients from controls in external clinical datasets with an AUC of 0.97 and in the Uppsala University Hospital datasets with an AUC of 0.99.

Discussion

CTVMs derived through deep learning, show potential for assessing and quantifying morphological features in hydrocephalus. Critically, these measures performed comparably to gold-standard neuroradiology assessments in distinguishing iNPH from healthy controls, even in the presence of intraventricular shunt catheters. Accordingly, such an approach may serve to improve the radiological evaluation of iNPH diagnosis/monitoring (i.e., treatment responses). Since CT is much more widely available than MRI, our results have considerable clinical impact.

Extensive T1-weighted MRI preprocessing improves generalizability of deep brain age prediction models

Brain age is an estimate of chronological age obtained from T1-weighted magnetic resonance images (T1w MRI), representing a straightforward diagnostic biomarker of brain aging and associated diseases. While the current best accuracy of brain age predictions on T1w MRIs of healthy subjects ranges from two to three years, comparing results across studies is challenging due to differences in the datasets, T1w preprocessing pipelines, and evaluation protocols used. This paper investigates the impact of T1w image preprocessing on the performance of four deep learning brain age models from recent literature. Four preprocessing pipelines, which differed in terms of registration transform, grayscale correction, and software implementation, were evaluated. The results showed that the choice of software or preprocessing steps could significantly affect the prediction error, with a maximum increase of 0.75 years in mean absolute error (MAE) for the same model and dataset. While grayscale correction had no significant impact on MAE, using affine rather than rigid registration to brain atlas statistically significantly improved MAE. Models trained on 3D images with isotropic 1mm3 resolution exhibited less sensitivity to the T1w preprocessing variations compared to 2D models or those trained on downsampled 3D images. Our findings indicate that extensive T1w preprocessing improves MAE, especially when predicting on a new dataset. This runs counter to prevailing research literature, which suggests that models trained on minimally preprocessed T1w scans are better suited for age predictions on MRIs from unseen scanners. We demonstrate that, irrespective of the model or T1w preprocessing used during training, applying some form of offset correction is essential to enable the model's performance to generalize effectively on datasets from unseen sites, regardless of whether they have undergone the same or different T1w preprocessing as the training set.