AI Networking Cookbook: Practical recipes for AI-assisted network automation and development
TL;DR
Comparison of different MRI modalities in predicting brain age and the relative importance of specific brain regions provides valuable insight into their relative predictive utility and their potential role in characterizing brain aging.
Abstract Brain age gap estimation (BrainAGE) is a promising imaging-derived biomarker of neurobiological ageing and disease risk, yet current approaches rely predominantly on T1-weighted structural MRI, overlooking functional vascular changes that may precede tissue damage and cognitive decline. Deep learning-derived cerebral blood volume (DeepCBV) maps, synthesized from non-contrast MRI, offer a scalable alternative to contrast-enhanced perfusion imaging by capturing vascular information relevant to early neurodegeneration. We developed a multimodal BrainAGE framework that combines predictions from two separate three-dimensional convolutional neural networks: one trained only on structural MRI scans and another trained only on DeepCBV maps generated by a pre-trained three-dimensional patch-based deep learning model. Each model was trained and validated on 2851 scans (1507 females) from 13 open-source datasets and was evaluated for concordance with mild cognitive impairment (MCI) and Alzheimer’s disease (AD) using 1233 subjects. The combined model achieved the most accurate brain age gap for cognitively normal (CN) controls, with a mean absolute error of 3.95 years (R2 = 0.943), outperforming models trained on MRI (mean absolute error = 4.10) or DeepCBV alone (mean absolute error = 4.49). Saliency maps revealed complementary modality contributions: MRI emphasized white matter and cortical atrophy, while DeepCBV highlighted vascular-rich and periventricular regions implicated in hypoperfusion and early cerebrovascular dysfunction, consistent with known patterns of normal ageing. Next, we observed that BrainAGE increased stepwise across diagnostic strata (CN < MCI < AD) and correlated with cognitive impairment (Clinical Dementia Rating Sum of Boxes ⍴ = 0.403; Mini-Mental State Examination ⍴ = −0.310). DeepCBV-based BrainAGE showed a particularly strong separation between stable versus progressive MCI (Mann–Whitney U = 2.177 × 104, P = 4.43 × 10−8), suggesting sensitivity to prodromal vascular changes that precede overt atrophy. Integrating structural MRI with deep learning-derived vascular measures substantially enhances BrainAGE estimation and improves sensitivity to MCI and Alzheimer’s disease progression, supporting its potential role in risk stratification, early detection and monitoring of therapeutic response. By enabling a functional-like assessment from routine MRI, this approach lowers barriers to multimodal evaluation and provides a clinically actionable biomarker for large-scale ageing and dementia studies.
Background Neuron loss is a hallmark of neurodegenerative diseases and leads to brain atrophy detectable with magnetic resonance imaging (MRI). Accurate prediction of future atrophy is valuable for research in Alzheimer's disease (AD) and related dementias. Objective This study aimed to predict annualized percentage changes in hippocampal, ventricular, and total gray matter (TGM) volumes in individuals ranging from cognitively normal to dementia, and to evaluate whether longitudinal MRI-derived change measures improve prediction performance compared with single-time-point MRI information. Methods Using elastic net regression, we compared baseline models based on single-timepoint MRI information with longitudinal models incorporating prior MRI-derived change measures. Both approaches were evaluated as MRI-only and MRI + risk-factor variants, with risk factors including age, sex, APOE4, and diagnostic status. Results In cross-validated analyses using the Alzheimer's Disease Neuroimaging Initiative (ADNI) cohort, the longitudinal MRI + risk-factor model performed best, yielding Pearson correlations of 0.62 for hippocampal atrophy, 0.51 for ventricular enlargement, and 0.41 for TGM atrophy. Longitudinal models consistently outperformed single time-point models, and adding risk factors improved predictive performance beyond MRI alone. External validation using the Australian Imaging, Biomarkers and Lifestyle cohort confirmed these findings. Predicted atrophy outperformed present-day regional volumes in identifying individuals progressing from normal cognition to MCI/dementia and from MCI to dementia. Conclusions MRI-derived longitudinal features enhance atrophy prediction, and predicted atrophy rates provide sensitive markers of future cognitive decline. These findings support the potential utility of predicted atrophy for cohort enrichment and therapeutic trial design.
Alzheimer’s disease (AD) is the leading cause of dementia and disproportionately affects women, who experience a higher lifetime risk and more rapid structural brain changes than men. Reliable imaging biomarkers are essential for detecting these changes, although their estimation may be influenced by voxel geometry, image resolution, and segmentation methodology. In this study, magnetic resonance imaging (MRI) scans from 40 female participants with AD obtained from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database were analyzed using the Computational Anatomy Toolbox 12 (CAT12), implemented within Statistical Parametric Mapping 12 (SPM12), and Materialise Mimics to identify structural biomarkers associated with neurodegeneration. Cortical thickness, gray matter (GM), white matter (WM), cerebrospinal fluid (CSF), and total intracranial volume (TIV) were quantified, while Brain Parenchymal Volume (BPV) and Brain Parenchymal Fraction (BPF) were calculated to assess global brain tissue preservation. The analyses demonstrated characteristic AD-related changes, including cortical thinning, GM loss, ventricular enlargement, and increased CSF volume. Significant correlations among cortical thickness, tissue volumes, BPV, and BPF further supported their complementary role in characterizing disease-related structural changes. Overall, these findings suggest that MRI-derived measures of cortical thickness, tissue volumes, BPV, and BPF provide useful structural biomarkers of AD and emphasize the importance of considering voxel geometry and segmentation methodology when evaluating neuroimaging biomarkers.
Three cognitive phenotypes with differential patterns of brain atrophy measures integrate neuropsychological testing with MRI measures into a clinically applicable framework that may help bridge the gap between structural imaging findings and everyday cognitive assessment in PwMS.
Mental illness is becoming increasingly common in everyday life, affecting a growing number of people. Functional Magnetic Resonance Imaging (fMRI) is an effective technique for detecting mental disorders. Resting-state fMRI (rs-fMRI) analyses spontaneous low-frequency oscillations in the Blood Oxygen Level Dependent signal to examine the functional architecture of the brain. fMRI and rs-fMRI data processing contribute to mapping neural activity in specific brain regions, enabling accurate localization of cognitive functions and improving our understanding of Functional Connectivity and interactions among brain regions. This paper focuses on the use of Machine Learning and Deep Learning techniques for fMRI analysis to identify brain features that may be associated with Bipolar Disorder (BD) and Alzheimer’s Disease (AD). The first step in the proposed approach is the preprocessing of fMRI and rs-fMRI data using two distinct methods: the first combines Statistical Parametric Mapping and the CONN toolbox for improved results, and the second focuses on the MELODIC software. The proposed approach begins with fMRI and rs-fMRI preprocessing using two pipelines: one integrating Statistical Parametric Mapping (SPM) with the CONN toolbox and another based on the MELODIC framework. The aim is to evaluate how preprocessing choices affect functional connectivity representations and classification performance when combined with classical Machine Learning (ML) methods, namely Support Vector Machines and Random Forests, and the Deep Learning (DL) model AlexNet. Given the altered brain connectivity patterns observed in Bipolar Disorder and Alzheimer’s Disease, these methods are employed to assess preprocessing-learning interactions across both disorders. Experimental results emphasize that Connbox-SPM preprocessing combined with AlexNet yields the best results for both mental disorders, with Bipolar Disorder reporting 78.2% accuracy and AD achieving 82.4% accuracy.
Abstract INTRODUCTION Brain‐age gap (BAG), the difference between predicted age and chronological age, is studied as a biomarker for the natural progression of neurodegeneration. The BAG captures brain atrophy as measured with structural magnetic resonance imaging (MRI). Electroencephalography (EEG) has also been explored for estimating the BAG (EEG‐BAG). However, studies showed mixed results for the EEG‐BAG including counterintuitive findings of younger predicted age in clinical populations, raising doubts about its utility as a clinical tool for mild cognitive impairment (MCI) and Alzheimer's disease (AD). METHODS This study critically examined brain‐age estimation from spectral EEG power as a common measure of brain activity in two of the largest public EEG datasets containing heterogeneous clinical cases alongside controls including MCI and AD. EEG recordings were analyzed from individuals with heterogeneous neurological conditions (n = 898, Temple University Hospital Abnormal EEG corpus [TUAB] data; n = 417 MCI & n = 311 dementia, Chung‐Ang University Hospital EEG [CAU] data) and controls (n = 1245, TUAB data; n = 459, CAU data). RESULTS We found that age‐prediction models trained on the reference population systematically underpredicted age in clinical conditions showing strong and systematic, age‐related differences in EEG power compared to controls. Data exploration and simulations revealed how diverging age‐related trends in specific EEG frequencies can account for a negative EEG‐BAG. DISCUSSION The utility of brain age as an interpretable biomarker relies on the observation from structural MRI that progressive neurodegeneration often broadly resembles aging. This assumption can be violated for functional assessments such as EEG spectral power related to different neurological and psychiatric conditions or medications. The sign of the BAG may therefore not be meaningfully interpreted as an individual aging metric, hence hampering its utility as an endpoint or biomarker in MCI and AD.
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.