The Alzheimer's Disease Neuroimaging Initiative 3: Continued innovation for clinical trial improvement

Association between diffusion MRI-based measures of neurite microstructure and risk of Alzheimer's disease

Early detection of Alzheimer's disease (AD) is crucial for intervention, but traditional MRI and cognitive assessments may miss pre-symptomatic changes. Advanced diffusion MRI (dMRI) methods, such as Neurite Orientation Dispersion and Density Imaging (NODDI), show promise in identifying early brain changes. We analyzed 65 cognitively unimpaired older adults (25 APOE-e4 carriers, 40 non-carriers) from the ADNI3 dataset. NODDI's neurite density index (NDI) and orientation dispersion index (ODI), volumetric MRI and cognition (MoCA) were analyzed in key brain regions like the hippocampus, fusiform gyrus, and entorhinal cortex. Statistical analyses included linear regression and t-tests, with FDR correction. NDI differed significantly between carriers and non-carriers and correlated with MoCA scores. ODI differed only in the CA1 hippocampal subfield. Volumetric MRI measures showed no group differences. Significant APOE-e4 group differences were observed in NDI for the left fusiform gyrus (β = 0.015, p = 0.02), right fusiform gyrus (β = 0.018, p = 0.02), left entorhinal cortex (β = 0.018, p = 0.04), right entorhinal cortex (β = 0.018, p = 0.03), left CA1 (β = 0.03, p = 0.02), and left CA2-3 (β = 0.03, p = 0.02). ODI differences were observed only in left CA1 (β = 0.037, p = 0.008). No volumetric measures differed significantly. MoCA correlated with NDI in bilateral entorhinal cortices (p = 0.001-0.05), left fusiform gyrus (p = 0.02), and right CA2-3 (p = 0.02). NODDI metrics, particularly NDI, could help detect early APOE-e4-related microstructural changes, while traditional volumetric MRI measures remain uninformative at early stages.

Polyconnectomic Scoring of Functional Connectivity Patterns Across Eight Neuropsychiatric and Three Neurodegenerative Disorders

BACKGROUND

Neuropsychiatric and neurodegenerative disorders involve diverse changes in brain functional connectivity. As an alternative to approaches that search for specific mosaic patterns of affected connections and networks, we used polyconnectomic scoring to quantify disorder-related whole-brain connectivity signatures into interpretable, personalized scores.

METHODS

The polyconnectomic score (PCS) measures the extent to which an individual's functional connectivity mirrors the whole-brain circuitry characteristics of a trait. We computed PCSs for 8 neuropsychiatric conditions (attention-deficit/hyperactivity disorder, anxiety-related disorders, autism spectrum disorder, obsessive-compulsive disorder, bipolar disorder, major depressive disorder, schizoaffective disorder, and schizophrenia) and 3 neurodegenerative conditions (Alzheimer's disease, frontotemporal dementia, and Parkinson's disease) across 22 datasets with resting-state functional magnetic resonance imaging data from 10,667 individuals (5325 patients, 5342 control participants). We also examined PCSs in 26,673 individuals from the population-based UK Biobank cohort.

RESULTS

PCSs were consistently higher in out-of-sample patients across 6 of the 8 neuropsychiatric and across all 3 investigated neurodegenerative disorders ([minimum, maximum]: area under the receiver operating characteristic curve = [0.55, 0.73], false discovery rate-corrected p [pFDR] = [1.8 × 10-16, 4.5 × 10-2]). Individuals with elevated PCS levels for neuropsychiatric conditions exhibited higher neuroticism (pFDR < 9.7 × 10-5), lower cognitive performance (pFDR < 5.3 × 10-5), and lower general well-being (pFDR < 9.7 × 10-4).

CONCLUSIONS

Our findings reveal generalizable whole-brain connectivity alterations in brain disorders. Polyconnectomic scoring effectively aggregates disorder-related signatures across the entire brain into an interpretable, participant-specific metric. A toolbox is provided for PCS computation.

Deep-Diffeomorphic Networks for Conditional Brain Templates

Deformable brain templates are an important tool in many neuroimaging analyses. Conditional templates (e.g., age-specific templates) have advantages over single population templates by enabling improved registration accuracy and capturing common processes in brain development and degeneration. Conventional methods require large, evenly spread cohorts to develop conditional templates, limiting their ability to create templates that could reflect richer combinations of clinical and demographic variables. More recent deep-learning methods, which can infer relationships in very high-dimensional spaces, open up the possibility of producing conditional templates that are jointly optimised for these richer sets of conditioning parameters. We have built on recent deep-learning template generation approaches using a diffeomorphic (topology-preserving) framework to create a purely geometric method of conditional template construction that learns diffeomorphisms between: (i) a global or group template and conditional templates, and (ii) conditional templates and individual brain scans. We evaluated our method, as well as other recent deep-learning approaches, on a data set of cognitively normal (CN) participants from the Alzheimer's Disease Neuroimaging Initiative (ADNI), using age as the conditioning parameter of interest. We assessed the effectiveness of these networks at capturing age-dependent anatomical differences. Our results demonstrate that while the assessed deep-learning methods have a number of strengths, they require further refinement to capture morphological changes in ageing brains with an acceptable degree of accuracy. The volumetric output of our method, and other recent deep-learning approaches, across four brain structures (grey matter, white matter, the lateral ventricles and the hippocampus), was measured and showed that although each of the methods captured some changes well, each method was unable to accurately track changes in all of the volumes. However, as our method is purely geometric, it was able to produce T1-weighted conditional templates with high spatial fidelity and with consistent topology as age varies, making these conditional templates advantageous for spatial registrations. The use of diffeomorphisms in these deep-learning methods represents an important strength of these approaches, as they can produce conditional templates that can be explicitly linked, geometrically, across age as well as to fixed, unconditional templates or brain atlases. The use of deep learning in conditional template generation provides a framework for creating templates for more complex sets of conditioning parameters, such as pathologies and demographic variables, in order to facilitate a broader application of conditional brain templates in neuroimaging studies. This can aid researchers and clinicians in their understanding of how brain structure changes over time and under various interventions, with the ultimate goal of improving the calibration of treatments and interventions in personalised medicine. The code to implement our conditional brain template network is available at: github.com/lwhitbread/deep-diff.

Precision Estimates of Longitudinal Brain Aging Capture Unexpected Individual Differences in One Year: Summary: A novel brain imaging method boosts precision to reveal variable brain aging trajectories

Longitudinal studies are required to measure individual differences in human brain aging, but they are difficult to estimate over short intervals because of measurement error. Using cluster scanning, an approach that reduces error by densely repeating rapid structural scans, we assessed brain aging in individuals across three longitudinal timepoints spaced across one year. Cluster scanning substantially improved the precision of individualized estimates, revealing previously undetectable individual differences in brain change. In just one year, expected differences in the rates of brain aging between younger and older individuals were evident, as were differences between cognitively unimpaired and impaired individuals. Each person's brain change trajectory was compared to modeled normative expectations from a large cohort of age-matched UK Biobank participants. Cognitively unimpaired older individuals variably revealed relative brain maintenance, unexpectedly rapid decline, and asymmetrical changes. These atypical brain aging trajectories were found across structures and verified in independent within-individual test and retest data. Cluster scanning promises to advance our understanding of the marked heterogeneity in brain aging by affording better short-term tracking of individual variability in structural change.

Lifespan reference curves for harmonizing multi-site regional brain white matter metrics from diffusion MRI

Age-related white matter (WM) microstructure maturation and decline occur throughout the human lifespan, complementing the process of gray matter development and degeneration. Here, we create normative lifespan reference curves for global and regional WM microstructure by harmonizing diffusion MRI (dMRI)-derived data from ten public datasets (N = 40,898 subjects; age: 3-95 years; 47.6% male). We tested three harmonization methods on regional diffusion tensor imaging (DTI) based fractional anisotropy (FA), a metric of WM microstructure, extracted using the ENIGMA-DTI pipeline. ComBat-GAM harmonization provided multi-study trajectories most consistent with known WM maturation peaks. Lifespan FA reference curves were validated with test-retest data and used to assess the effect of the ApoE4 risk factor for dementia in WM across the lifespan. We found significant associations between ApoE4 and FA in WM regions associated with neurodegenerative disease even in healthy individuals across the lifespan, with regional age-by-genotype interactions. Our lifespan reference curves and tools to harmonize new dMRI data to the curves are publicly available as eHarmonize ( https://github.com/ahzhu/eharmonize ).

Probabilistic Mapping and Automated Segmentation of Human Brainstem White Matter Bundles

Brainstem white matter bundles are essential conduits for neural signaling involved in modulation of vital functions ranging from homeostasis to human consciousness. Their architecture forms the anatomic basis for brainstem connectomics, subcortical mesoscale circuit models, and deep brain navigation tools. However, their small size and complex morphology compared to cerebral white matter structures makes mapping and segmentation challenging in neuroimaging. This results in a near absence of automated brainstem white matter tracing methods. We leverage diffusion MRI tractography to create BrainStem Bundle Tool (BSBT), which segments eight key white matter bundles in the rostral brainstem. BSBT performs automated segmentation on a custom probabilistic fiber map generated from tractography with a convolutional neural network architecture tailored for detection of small structures. We demonstrate BSBTs robustness across diffusion MRI acquisition protocols through validation on healthy subject in vivo scans and ex vivo scans of brain specimens with corresponding histology. Using BSBT, we reveal distinct brainstem white matter bundle alterations in Alzheimer's disease, Parkinson's disease, and acute traumatic brain injury cohorts through tract-based analysis and classification tasks. Finally, we provide proof-of-principle evidence supporting the prognostic utility of BSBT in a longitudinal analysis of coma recovery. BSBT creates opportunities to automatically map brainstem white matter in large imaging cohorts and investigate its role in a broad spectrum of neurological disorders.

Brain activation in older adults during odor identification is related to ApoE, t-tau/Aβ1-42, and hippocampal volume

Despite altered odor identification preceding and predicting Alzheimer's disease (AD) cognitive decline, an inadequate understanding of how AD pathology affects odor memory functions limits its use as a preclinical biomarker. Multivariate linear regression was applied to whole-brain blood-oxygen-level-dependent (BOLD) activations during odor identification task (OID) responses in older adults without dementia (N = 36, 44.4 % ε4 carriers, MAge= 76.61). Apolipoprotein-E ε4 allele status, cerebrospinal fluid levels of total-tau to Amyloid-β1-42, and MRI-derived hippocampal volume measures were used as predictors. The predictors described significant BOLD variation in regions that are associated with necessary OID functions and affected by AD neurodegeneration during OID responses; moreover, all predictors were associated with significant (P < .001) negative BOLD effects in essential task regions during at least one response condition. This evidence suggests significant pathological effects of AD biomarkers on OID-response neural activity in older adults without dementia and should motivate future combined-biomarker investigations of OID functions in preclinical populations.

A lifespan-generalizable skull-stripping model for magnetic resonance images that leverages prior knowledge from brain atlases

In magnetic resonance imaging of the brain, an imaging-preprocessing step removes the skull and other non-brain tissue from the images. But methods for such a skull-stripping process often struggle with large data heterogeneity across medical sites and with dynamic changes in tissue contrast across lifespans. Here we report a skull-stripping model for magnetic resonance images that generalizes across lifespans by leveraging personalized priors from brain atlases. The model consists of a brain extraction module that provides an initial estimation of the brain tissue on an image, and a registration module that derives a personalized prior from an age-specific atlas. The model is substantially more accurate than state-of-the-art skull-stripping methods, as we show with a large and diverse dataset of 21,334 lifespans acquired from 18 sites with various imaging protocols and scanners, and it generates naturally consistent and seamless lifespan changes in brain volume, faithfully charting the underlying biological processes of brain development and ageing.

Multimodal spatial gradients to explain regional susceptibility to fibrillar tau in Alzheimer's disease

INTRODUCTION

In Alzheimer's disease (AD), fibrillar tau gradually progresses from initial seed to larger brain area. However, those brain properties underlying the region-dependent susceptibility to tau accumulation remain unclear.

METHODS

We constructed multimodal spatial gradients to characterize molecular properties and connectomic architecture. A predictive model for regional tau deposition was developed by integrating embeddings in the principal gradients of global connectome gradients with gene expression, neurotransmitters, myelin, and amyloid-beta. The model was trained on amyloid-beta-positive participants from Alzheimer's Disease Neuroimaging Initiative (ADNI) and externally validated in independent datasets.

RESULTS

The combination of gradients explained up to 77.7% of cross-sectional and 77.3% of longitudinal inter-regional variance of tau deposition. Gene set enrichment analysis of a major gene expression gradient points to synaptic transmission to confer increased susceptibility to tau.

DISCUSSION

Our findings reveal a spatially heterogeneous molecular landscape shaping regional susceptibility to tau deposition, presenting a powerful system-level explanatory model of tau pathology in AD.

HIGHLIGHTS

Spatial gradients of fundamental molecular brain properties associated with tau pathology. The explanatory power showed high consistency across studies. Genetic analyses suggested that synapse expression plays a vital role in tau accumulation.

A foundation model for enhancing magnetic resonance images and downstream segmentation, registration and diagnostic tasks

In structural magnetic resonance (MR) imaging, motion artefacts, low resolution, imaging noise and variability in acquisition protocols frequently degrade image quality and confound downstream analyses. Here we report a foundation model for the motion correction, resolution enhancement, denoising and harmonization of MR images. Specifically, we trained a tissue-classification neural network to predict tissue labels, which are then leveraged by a 'tissue-aware' enhancement network to generate high-quality MR images. We validated the model's effectiveness on a large and diverse dataset comprising 2,448 deliberately corrupted images and 10,963 images spanning a wide age range (from foetuses to elderly individuals) acquired using a variety of clinical scanners across 19 public datasets. The model consistently outperformed state-of-the-art algorithms in improving the quality of MR images, handling pathological brains with multiple sclerosis or gliomas, generating 7-T-like images from 3 T scans and harmonizing images acquired from different scanners. The high-quality, high-resolution and harmonized images generated by the model can be used to enhance the performance of models for tissue segmentation, registration, diagnosis and other downstream tasks.

Establishing Regional Aβ Cutoffs and Exploring Subgroup Prevalence Across Cognitive Stages Using BeauBrain Amylo®

Background and Purpose

Amyloid-beta (Aβ) plaques are key in Alzheimer's disease (AD), with Aβ positron emission tomography imaging enabling non-invasive quantification. To address regional Aβ deposition, we developed regional Centiloid scales (rdcCL) and commercialized them through the computed tomography (CT)-based BeauBrain Amylo platform, eliminating the need for three-dimensional T1 magnetic resonance imaging (MRI).

Objective

We aimed to establish robust regional Aβ cutoffs using the commercialized BeauBrain Amylo platform and to explore the prevalence of subgroups defined by global, regional, and striatal Aβ cutoffs across cognitive stages.

Methods

We included 2,428 individuals recruited from the Korea-Registries to Overcome Dementia and Accelerate Dementia Research project. We calculated regional Aβ cutoffs using Gaussian Mixture Modeling. Participants were classified into subgroups based on global, regional, and striatal Aβ positivity across cognitive stages (cognitively unimpaired [CU], mild cognitive impairment, and dementia of the Alzheimer's type).

Results

MRI-based and CT-based global Aβ cutoffs were highly comparable and consistent with previously reported Centiloid values. Regional cutoffs revealed both similarities and differences between MRI- and CT-based methods, reflecting modality-specific segmentation processes. Subgroups such as global(-)regional(+) were more frequent in non-dementia stages, while global(+)striatal(-) was primarily observed in CU individuals.

Conclusions

Our study established robust regional Aβ cutoffs using a CT-based rdcCL method and demonstrated its clinical utility in classifying amyloid subgroups across cognitive stages. These findings highlight the importance of regional Aβ quantification in understanding amyloid pathology and its implications for biomarker-guided diagnosis and treatment in AD.

Assessing measurement consistency of a diffusion tensor imaging (DTI) quality control (QC) anisotropy phantom

OBJECTIVES

We evaluated a quality control (QC) phantom designed to mimic diffusion characteristics and white matter fiber tracts in the brain. We hypothesized that acquisition of diffusion tensor imaging (DTI) data on different vendors and over multiple repeated measures would not contribute to significant variability in calculated diffusion tensor scalar metrics such as fractional anisotropy (FA) and mean diffusivity (MD).

MATERIALS AND METHODS

The DTI QC phantom was scanned using a 32-direction DTI sequence on General Electric (GE), Siemens, and Philips 3 Tesla scanners. Motion probing gradients (MPGs) were investigated as a source of variance in our statistical design, and data were acquired on GE and Siemens scanners using GE, Siemens, and Philips vendor MPGs for 32 directions. In total, 8 repeated scans were made for each GE/Siemens combination of vendor and MPGs with 8 repeated scans on a Philips machine using its stock DTI sequence. Data were analyzed using 2-way ANOVAs to investigate repeat scan and vendor variances and 3-way ANOVAs with repeat, MPG, and vendor as factors.

RESULTS

No statistical differences (i.e., P > 0.05) were found in any DTI scalar metrics (FA, MD) or for any factor, suggesting system constancy across imaging platforms and the specified phantom's reliability and reproducibility across vendors and conditions.

DISCUSSION

A DTI QC phantom demonstrates that DTI measurements maintain their consistency across different MRI systems and can contribute to a standard that is more reliable for quantitative MRI analyses.

Class balancing diversity multimodal ensemble for Alzheimer's disease diagnosis and early detection

Alzheimer's disease (AD) poses significant global health challenges due to its increasing prevalence and associated societal costs. Early detection and diagnosis of AD are critical for delaying progression and improving patient outcomes. Traditional diagnostic methods and single-modality data often fall short in identifying early-stage AD and distinguishing it from Mild Cognitive Impairment (MCI). This study addresses these challenges by introducing a novel approach: multImodal enseMble via class BALancing diversity for iMbalancEd Data (IMBALMED). IMBALMED integrates multimodal data from the Alzheimer's Disease Neuroimaging Initiative database, including clinical assessments, neuroimaging phenotypes, biospecimen, and subject characteristics data. It employs a new ensemble of model classifiers, designed specifically for this framework, which combines eight distinct families of learning paradigms trained with diverse class balancing techniques to overcome class imbalance and enhance model accuracy. We evaluate IMBALMED on two diagnostic tasks (binary and ternary classification) and four binary early detection tasks (at 12, 24, 36, and 48 months), comparing its performance with state-of-the-art algorithms and an unbalanced dataset method. To further validate the proposed model and ensure genuine generalization to real-world scenarios, we conducted an external validation experiment using data from the most recent phase of the ADNI dataset. IMBALMED demonstrates superior diagnostic accuracy and predictive performance in both binary and ternary classification tasks, significantly improving early detection of MCI at a 48-month time point and showing excellent generalizability in the 12-month task during external validation. The method shows improved classification performance and robustness, offering a promising solution for early detection and management of AD.

An imaging and genetic-based deep learning network for Alzheimer's disease diagnosis

Conventional computer-aided diagnostic techniques for Alzheimer's disease (AD) predominantly rely on magnetic resonance imaging (MRI) in isolation. Genetic imaging methods, by establishing the link between genes and brain structures in disease progression, facilitate early prediction of AD development. While deep learning methods based on MRI have demonstrated promising results for early AD diagnosis, the limited dataset size has led most AD studies to lean on statistical approaches within the realm of imaging genetics. Existing deep-learning approaches typically utilize pre-defined regions of interest and risk variants from known susceptibility genes, employing relatively straightforward feature fusion methods that fail to fully capture the relationship between images and genes. To address these limitations, we proposed a multi-modal deep learning classification network based on MRI and single nucleotide polymorphism (SNP) data for AD diagnosis and mild cognitive impairment (MCI) progression prediction. Our model leveraged a convolutional neural network (CNN) to extract whole-brain structural features, a Transformer network to capture genetic features, and employed a cross-transformer-based network for comprehensive feature fusion. Furthermore, we incorporated an attention-map-based interpretability method to analyze and elucidate the structural and risk variants associated with AD and their interrelationships. The proposed model was trained and evaluated using 1,541 subjects from the ADNI database. Experimental results underscored the superior performance of our model in effectively integrating and leveraging information from both modalities, thus enhancing the accuracy of AD diagnosis and prediction.

Stages prediction of Alzheimer's disease with shallow 2D and 3D CNNs from intelligently selected neuroimaging data

Detection of Alzheimer's Disease (AD) is critical for successful diagnosis and treatment, involving the common practice of screening for Mild Cognitive Impairment (MCI). However, the progressive nature of AD makes it challenging to identify its causal factors. Modern diagnostic workflows for AD use cognitive tests, neurological examinations, and biomarker-based methods, e.g., cerebrospinal fluid (CSF) analysis and positron emission tomography (PET) imaging. While these methods are effective, non-invasive imaging techniques like Magnetic Resonance Imaging (MRI) are gaining importance. Deep Learning (DL) approaches for evaluating alterations in brain structure have focused on combining MRI and Convolutional Neural Networks (CNNs) within the spatial architecture of DL. This combination has garnered significant research interest due to its remarkable effectiveness in automating feature extraction across various multilayer perceptron models. Despite this, MRI's noisy and multidimensional nature requires an intelligent preprocessing pipeline for effective disease prediction. Our study aims to detect different stages of AD from the multidimensional neuroimaging data obtained through MRI scans using 2D and 3D CNN architectures. The proposed preprocessing pipeline comprises skull stripping, spatial normalization, and smoothing. It is followed by a novel and efficient pixel count-based frame selection and cropping approach, which renders a notable dimension reduction. Furthermore, the learnable resizer method is applied to enhance the image quality while resizing the data. Finally, the proposed shallow 2D and 3D CNN architectures extract spatio-temporal attributes from the segmented MRI data. Furthermore, we merged both the CNNs for further comparative analysis. Notably, 2D CNN achieved a maximum accuracy of 93%, while 3D CNN reported the highest accuracy of 96.5%.

Comparison of Deep Learning and Traditional Machine Learning Models for Predicting Mild Cognitive Impairment Using Plasma Proteomic Biomarkers

Mild cognitive impairment (MCI) is a clinical condition characterized by a decline in cognitive ability and progression of cognitive impairment. It is often considered a transitional stage between normal aging and Alzheimer's disease (AD). This study aimed to compare deep learning (DL) and traditional machine learning (ML) methods in predicting MCI using plasma proteomic biomarkers. A total of 239 adults were selected from the Alzheimer's Disease Neuroimaging Initiative (ADNI) cohort along with a pool of 146 plasma proteomic biomarkers. We evaluated seven traditional ML models (support vector machines (SVMs), logistic regression (LR), naïve Bayes (NB), random forest (RF), k-nearest neighbor (KNN), gradient boosting machine (GBM), and extreme gradient boosting (XGBoost)) and six variations of a deep neural network (DNN) model-the DL model in the H2O package. Least Absolute Shrinkage and Selection Operator (LASSO) selected 35 proteomic biomarkers from the pool. Based on grid search, the DNN model with an activation function of "Rectifier With Dropout" with 2 layers and 32 of 35 selected proteomic biomarkers revealed the best model with the highest accuracy of 0.995 and an F1 Score of 0.996, while among seven traditional ML methods, XGBoost was the best with an accuracy of 0.986 and an F1 Score of 0.985. Several biomarkers were correlated with the APOE-ε4 genotype, polygenic hazard score (PHS), and three clinical cerebrospinal fluid biomarkers (Aβ42, tTau, and pTau). Bioinformatics analysis using Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) revealed several molecular functions and pathways associated with the selected biomarkers, including cytokine-cytokine receptor interaction, cholesterol metabolism, and regulation of lipid localization. The results showed that the DL model may represent a promising tool in the prediction of MCI. These plasma proteomic biomarkers may help with early diagnosis, prognostic risk stratification, and early treatment interventions for individuals at risk for MCI.

Towards Clinical Diagnoses: Classifying Alzheimer's Disease Using Single fMRI, Small Datasets, and Transfer Learning

PURPOSE

Deep learning and functional magnetic resonance imaging (fMRI) are two unique methodologies that can be combined to diagnose Alzheimer's disease (AD). Multiple studies have harnessed these methods to diagnose AD with high accuracy. However, there are difficulties in adapting this research to real-world diagnoses. For example, the two key issues of data availability and model usability limit clinical applications. These two areas are concerned with problems of accessibility, generalizability, and methodology that may limit model adoption. For example, fMRI deep learning models require a large amount of training data, which is not widely available. Contemporary models are also not typically formatted for clinical data or created for use by non-specialized populations. In this study, we develop a deep-learning fMRI pipeline that addresses some of these issues.

METHOD

We use transfer learning to address problems with data availability. We also use semi-automated and single-image techniques (i.e., one fMRI volume per participant) to make a model that is usable for non-specialized populations. Our model was initially trained on 524 participants from the Autism Brain Imaging Data Exchange (ABIDE; Autism and controls). Our model was then transferred and fine-tuned to a small sample of 64 participants from the Alzheimer's Disease Neuroimaging Initiative (ADNI; AD and controls).

FINDINGS AND CONCLUSION

This transfer learning model achieved an AD classification accuracy of 77% and outperformed the same model without transfer learning by approximately 30%. Accordingly, our model showed that small AD samples can be accurately classified in a clinically friendly manner.

Bridging the gap to clinical use: A systematic review on TMS-EEG test-retest reliability

BACKGROUND

Transcranial magnetic stimulation (TMS) combined with electroencephalography (EEG) can provide insight on cortical excitability and brain circuits. TMS-evoked potentials (TEPs) are phase-locked waveforms reflecting neural activity, with potential applications in psychiatry and neurology. However, the reliability of TEPs remains underexplored, hindering clinical standardization. This systematic review evaluates TEP reliability, focusing on commonly used measures and assessments.

METHODS

A systematic review was conducted on PubMed for studies from 2002 to October 10, 2024, using keywords combining TMS, EEG, and reliability terms. Systematic reviews and non-English articles were excluded.

RESULTS

Eighteen studies met inclusion criteria, mostly assessing young, healthy populations. Late TEP components demonstrated high relative reliability, while early components exhibited lower reliability and variability across sessions. Analytical methods like the intraclass and concordance correlation coefficients, and Pearson's correlations consistently favored late TEPs.

DISCUSSION

Late TEPs exhibit higher reliability, while early components require further research. TMS artifacts complicate interpretation, in both late and early responses. Formal reliability assessments, standardized protocols, and diverse populations are essential for advancing TEP reliability for clinical application.

CONCLUSIONS

A more comprehensive reliability assessments is needed before the implementation of clinical applications.

AN INTERPRETABLE GENERATIVE MULTIMODAL NEUROIMAGING-GENOMICS FRAMEWORK FOR DECODING ALZHEIMER'S DISEASE

Objective

Alzheimer's disease (AD) is the most prevalent form of dementia worldwide, encompassing a prodromal stage known as Mild Cognitive Impairment (MCI), where patients may either progress to AD or remain stable. The objective of the work was to capture structural and functional modulations of brain structure and function relying on multimodal MRI data and Single Nucleotide Polymorphisms, also in case of missing views, with the twofold goal of classifying AD patients versus healthy controls and detecting MCI converters.

Approach

We propose a multimodal DL-based classification framework where a generative module employing Cycle Generative Adversarial Networks was introduced in the latent space for imputing missing data (a common issue of multimodal approaches). Explainable AI method was then used to extract input features' relevance allowing for post-hoc validation and enhancing the interpretability of the learned representations.

Main results

Experimental results on two tasks, AD detection and MCI conversion, showed that our framework reached competitive performance in the state-of-the-art with an accuracy of 0.926 ± 0.02 and 0.711 ± 0.01 in the two tasks, respectively. The interpretability analysis revealed gray matter modulations in cortical and subcortical brain areas typically associated with AD. Moreover, impairments in sensory-motor and visual resting state networks along the disease continuum, as well as genetic mutations defining biological processes linked to endocytosis, amyloid-beta, and cholesterol, were identified.

Significance

Our integrative and interpretable DL approach shows promising performance for AD detection and MCI prediction while shedding light on important biological insights.

Parallel changes in cognition, neuropsychiatric symptoms, and amyloid in cognitively unimpaired older adults and those with mild cognitive impairment

INTRODUCTION

Alzheimer's disease (AD) diagnosis centers on cognitive impairment despite other early indicators like neuropsychiatric symptoms (NPSs) and amyloid beta (Aβ) accumulation. This study examined how cognition, NPS, and Aβ changes are interrelated over time in individuals without dementia.

METHODS

Participants were 1247 individuals from the Alzheimer's Disease Neuroimaging Initiative (ADNI)-2 and -3 cohorts with at least 48 months of follow-up. Cognitive domains were assessed via ADNI composite measures, NPS via the neuropsychiatric inventory, and Aβ via standardized uptake value ratio (SUVR) composite scores. Co-occurring changes were evaluated with parallel process models.

RESULTS

NPS was longitudinally associated with performance in each cognitive domain. Negative baseline Aβ-cognition associations were observed in three cognitive domains. No Aβ-NPS associations were observed.

DISCUSSION

This study demonstrated strong longitudinal relationships between NPS and cognition in preclinical and prodromal stages of AD. Future studies should incorporate NPS into models of disease trajectories to improve early detection and prediction of disease progression.

HIGHLIGHTS

Co-occurring changes in Aβ, cognition, and neuropsychiatric symptoms are understudied. We found relationships between neuropsychiatric symptoms and cognition. We found baseline, but not longitudinal, Aβ and cognition associations. Changes in neuropsychiatric symptoms should be included in early detection models of ADRD.

Biomarker treatment effects in two phase 3 trials of gantenerumab

INTRODUCTION

We report biomarker treatment effects in the GRADUATE I and II phase 3 studies of gantenerumab in early Alzheimer's disease (AD).

METHODS

Amyloid and tau positron emission tomography (PET), volumetric magnetic resonance imaging (vMRI), cerebrospinal fluid (CSF), and plasma biomarkers used to assess gantenerumab treatment related changes on neuropathology, neurodegeneration, and neuroinflammation over 116 weeks.

RESULTS

Gantenerumab reduced amyloid PET load, CSF biomarkers of amyloid beta (Aβ)40, total tau (t-tau), phosphorylated tau 181 (p-tau181), neurogranin, S100 calcium-binding protein B (S100B), neurofilament light (NfL), alpha-synuclein (α-syn), neuronal pentraxin-2 (NPTX2), and plasma biomarkers of t-tau, p-tau181, p-tau217, and glial fibrillary acidic protein (GFAP) while increasing plasma Aβ40, Aβ42. vMRI showed increased reduction in whole brain volume and increased ventricular expansion, while hippocampal volume was unaffected. Tau PET showed no treatment effect.

DISCUSSION

Robust treatment effects were observed for multiple biomarkers in GRADUATE I and II. Comparison across anti-amyloid antibodies indicates utility of p-tau and GFAP as biomarkers of amyloid plaque removal while NfL and tau PET seem unsuitable as consistent indicators of clinical efficacy. vMRI might be confounded by non-neurodegenerative brain volume changes. TRIAL REGISTRATION NUMBER (CLINICALTRIALS.GOV IDENTIFIER): NCT03444870 and NCT03443973.

HIGHLIGHTS

Gantenerumab significantly reduced brain amyloid load. Tau positron emission tomography showed no treatment effect in a small subset of participants. Volumetric magnetic resonance imaging showed increased whole brain volume reduction under treatment while hippocampal volume was unaffected. Robust treatment effects on cerebrospinal fluid and plasma biomarkers were found, despite lack of clinical efficacy.

Characterization of cognitive decline in long-duration type 1 diabetes by cognitive, neuroimaging, and pathological examinations

BACKGROUNDWe aimed to characterize factors associated with the under-studied complication of cognitive decline in aging people with long-duration type 1 diabetes (T1D).METHODSJoslin "Medalists" (n = 222; T1D ≥ 50 years) underwent cognitive testing. Medalists (n = 52) and age-matched nondiabetic controls (n = 20) underwent neuro- and retinal imaging. Brain pathology (n = 26) was examined. Relationships among clinical, cognitive, and neuroimaging parameters were evaluated.RESULTSCompared with controls, Medalists had worse psychomotor function and recall, which associated with female sex, lower visual acuity, reduced physical activity, longer diabetes duration, and higher inflammatory cytokines. On neuroimaging, compared with controls, Medalists had significantly lower total and regional brain volumes, equivalent to 9 years of accelerated aging, but small vessel disease markers did not differ. Reduced brain volumes associated with female sex, reduced psychomotor function, worse visual acuity, longer diabetes duration, and higher inflammation, but not with glycemic control. Worse cognitive function, lower brain volumes, and diabetic retinopathy correlated with thinning of the outer retinal nuclear layer. Worse baseline visual acuity associated with declining psychomotor function in longitudinal analysis. Brain volume mediated the association between visual acuity and psychomotor function by 57%. Brain pathologies showed decreased volumes, but predominantly mild vascular or Alzheimer's-related pathology.CONCLUSION To our knowledge, this is the first comprehensive study of cognitive function, neuroimaging, and pathology in aging T1D individuals demonstrated that cognitive decline was related to parenchymal rather than neurovascular abnormalities, unlike type 2 diabetes, suggestive of accelerated aging in T1D. Improving visual acuity could perhaps be an important preventive measure against cognitive decline in people with T1D.FUNDINGThe Beatson Foundation, NIH/NIDDK grants 3P30DK036836-34S1 and P30DK036836-37, and Mary Iacocca fellowships.

Tau Burden is Best Captured by Magnitude and Extent: Tau-MaX as a Measure of Global Tau

Tau exhibits change in both spatial extent and density of pathology along the Alzheimer's disease (AD) spectrum with each aspect contributing to the overall burden of pathological tau. Nevertheless, studies using Tau PET have measured either magnitude using standardized uptake value ratios (SUVRs) or extent using number of Tau+ regions. We hypothesized that combining these two dimensions into a single measure of Magnitude and eXtent, Tau-MaX, would provide improved quantification of global tau burden as well as allowing for a region-agnostic measure of global tau burden that does not require a pre-specified region of interest (ROI) or meta-ROI. To test this hypothesis, we analyzed 18F-flortaucipir PET scans from local and national consortium data (n=1077 participants total) and used Gaussian-mixture models for data from 64 brain regions, to define both tau positivity and magnitude. We examined cross-sectional and longitudinal change in Tau-MaX across the Alzheimer's disease (AD) spectrum and compared the association of Tau-MaX, magnitude, and extent with plasma p-tau217 and global cognition. We also compared Tau-MaX using a global, region-agnostic approach to temporal lobe or Braak stage meta-ROIs. Whereas separate assessments of extent and magnitude across the disease spectrum found earlier increases in Tau spatial extent and later increases in magnitude, Tau-MaX was able to dynamically capture this shift demonstrating a stronger association with extent in the preclinical stage and a stronger association with magnitude in clinical stages. Global Tau-MaX differed between disease stages cross-sectionally and changed over time in all stages of disease. Further, Tau-MaX significantly improved associations with plasma p-tau217 and global cognition compared to magnitude or extent alone. Finally, global measures of Tau-MaX performed similarly to meta-ROI measures of Tau-MaX. Together, these findings indicate that combining magnitude and extent provides a robust measure of global tau burden that changes throughout the disease course and is associated with blood-based biomarkers and cognition. This measure may be of particular use for disease staging, as well as serving as an outcome measure to monitor response to therapeutic intervention.

POINTER Study Group and for the Alzheimer's Disease Neuroimaging Initiative. Positron emission tomography harmonization in the Alzheimer's Disease Neuroimaging Initiative: A scalable and rigorous approach to multisite amyloid and tau quantification

INTRODUCTION

A key goal of the Alzheimer's Disease NeuroImaging Initiative (ADNI) positron emission tomography (PET) Core is to harmonize quantification of β-amyloid (Aβ) and tau PET image data across multiple scanners and tracers.

METHODS

We developed an analysis pipeline (Berkeley PET Imaging Pipeline, B-PIP) for ADNI Aβ and tau PET images and applied it to PET data from other multisite studies. Steps include image pre-processing, refacing, magnetic resonance imaging (MRI)/PET co-registration, visual quality control (QC), quantification of tracer uptake, and standardization of Aβ and tau standardized uptake value ratios (SUVrs) across tracers.

RESULTS

Measurements from 10,105 cross-sectional and longitudinal Aβ and tau PET scans acquired in several studies between 2010 and 2024 can be processed, harmonized, and directly merged across tracers and cohorts.

DISCUSSION

The B-PIP developed in ADNI is a scalable image harmonization approach used in several observational studies and clinical trials that facilitates rigorous Aβ and tau PET quantification and data sharing.

HIGHLIGHTS

Quantitative results from ADNI Aβ and tau PET data are generated using a rigorous, scalable image processing pipeline This pipeline has been applied to PET data from several other large, multisite studies and trials Quantitative outcomes are harmonizable across studies and are shared with the scientific community.

Overview of Alzheimer's Disease Neuroimaging Initiative and future clinical trials

The overall goal of the Alzheimer's Disease Neuroimaging Initiative (ADNI) is to optimize and validate biomarkers for clinical trials while sharing all data and biofluid samples with the global scientific community. ADNI has been instrumental in standardizing and validating amyloid beta (Aβ) and tau positron emission tomography (PET) imaging. ADNI data were used for the US Food and Drug Administration (FDA) approval of the Fujirebio and Roche Elecsys cerebrospinal fluid diagnostic tests. Additionally, ADNI provided data for the trials of the FDA-approved treatments aducanumab, lecanemab, and donanemab. More than 6000 scientific papers have been published using ADNI data, reflecting ADNI's promotion of open science and data sharing. Despite its enormous success, ADNI has some limitations, particularly in generalizing its data and findings to the entire US/Canadian population. This introduction provides a historical overview of ADNI and highlights its significant accomplishments and future vision to pioneer "the clinical trial of the future" focusing on demographic inclusivity. HIGHLIGHTS: The Alzheimer's Disease Neuroimaging Initiative (ADNI) introduced a novel model for public-private partnerships and data sharing. It successfully validated amyloid and Tau PET imaging, as well as CSF and plasma biomarkers, for diagnosing Alzheimer's disease. ADNI generated and disseminated vital data for designing AD clinical trials.

Improved Correlation of 18F-Flortaucipir PET SUVRs and Clinical Stages in the Alzheimer Disease Continuum with the MUBADA/PERSI-Based Analysis

The Alzheimer disease (AD) continuum is a neurodegenerative disorder with cognitive decline and pathologic changes. Tau PET imaging can detect tau pathology, and 18F-flortaucipir PET imaging is expected to visualize progression through the stages of AD, for which quantitative assessment is essential. Two measurement methods, statistically defined multiblock barycentric discriminant analysis (MUBADA)/parametric estimation of reference signal intensity (PERSI) and anatomically defined tau meta-volume of interest (VOI)/cerebellar gray matter (CGM) for SUV ratio (SUVR), were compared in this study to assess their relationship to AD clinical stage using 2 open multicenter PET databases. Methods: Data were selected for 106 cases from 2 databases, AMED Preclinical AD study (AMED-PRE) (n = 15) and Alzheimer Disease Neuroimaging Initiative 3 (n = 91). The data of the participants were categorized into 4 groups based on the clinical criteria. Tau PET imaging was conducted using 18F-flortaucipir, and the 2 SUVR measurement methods, MUBADA/PERSI and tau meta-VOI/CGM, were compared among different clinical categories: amyloid-negative cognitively normal, preclinical AD, amyloid-negative mild cognitive impairment (MCI), and amyloid-positive MCI. Results: Significant differences were found between cognitively normal and preclinical AD, as well as between cognitively normal and amyloid-positive MCI and between amyloid-negative MCI and -positive MCI in SUVR derived by MUBADA/PERSI, whereas SUVR by tau meta-VOI/CGM did not provide significant differences between any pair. The tau meta-VOI/CGM method consistently provided higher SUVRs and larger individual variations than MUBADA/PERSI, with a mean SUVR difference of 0.136 for the studied databases. Conclusion: MUBADA/PERSI provided the SUVR of 18F-flortaucipir uptake with better association with the clinical severity of the AD continuum and with smaller variability. The results support the usefulness of MUBADA/PERSI as a quantitative measure of 18F-flortaucipir uptake in multicenter studies using different PET systems and scanning methods. However, limitations of the study include the small sample size and the unbalanced distribution among clinical categories in the AMED Preclinical AD study database.

Pathway enrichment in genome-wide analysis of longitudinal Alzheimer's disease biomarker endophenotypes

INTRODUCTION

The genetic pathways that influence longitudinal heterogeneous changes in Alzheimer's disease (AD) may provide insight into disease mechanisms and potential therapeutic targets.

METHODS

Longitudinal endophenotypes from the Alzheimer's Disease Neuroimaging Initiative (ADNI) representing amyloid, tau, neurodegeneration (A/T/N), and cognition were selected. Genome-wide association analysis was performed using a linear mixed model (LMM) approach, followed by gene and pathway enrichment with significant and functionally relevant SNPs.

RESULTS

A total of 33 and 19 statistically significant pathways were identified associating with the intercept and longitudinal trajectory, respectively. The longitudinal intercept pathways represent eight groups: immune, metabolic, cell growth and survival, DNA maintenance, neuronal signaling, RAS/MAPK/ERK signaling pathways, vesicle and lysosomal transport, and transcription modification. Longitudinal trajectory pathways represented six groups: Immune, metabolic, cell signaling, cytoskeleton, and glycosylation.

DISCUSSION

Longitudinal enrichment identified pathways that uniquely associate with trajectories of key AD biomarkers and cognition, providing new insight into AD course-related mechanisms and potential new therapeutic targets.

HIGHLIGHTS

A systematic genome-wide analysis with longitudinal AD biomarker endophenotypes was performed. Enriched pathways were identified with functionally derived SNP to gene analysis. Fifty-two pathways were associated with longitudinal trajectory and intercept. Many of the identified pathways are specific steps in larger pathways implicated in AD. The identified pathways may provide therapeutic targets and areas for further study.

Cerebrovascular markers of WMH and infarcts in ADNI: A historical perspective and future directions

White matter hyperintensities (WMH) and infarcts found on magnetic resonance imaging (MR infarcts) are common biomarkers of cerebrovascular disease. In this review, we summarize the methods, publications, and conclusions stemming from the Alzheimer's Disease Neuroimaging Initiative (ADNI) related to these measures. We combine analysis of WMH and MR infarct data from across the three main ADNI cohorts with a review of existing literature discussing new methodologies and scientific findings derived from these data. Although ADNI inclusion criteria were designed to minimize vascular risk factors and disease, data across all the ADNI cohorts found consistent trends of increasing WMH volumes associated with advancing age, female sex, and cognitive impairment. ADNI, initially proposed as a study to investigate biomarkers of AD pathology, has also helped elucidate the impact of asymptomatic cerebrovascular brain injury on cognition within a cohort relatively free of vascular disease. Future ADNI work will emphasize additional vascular biomarkers. HIGHLIGHTS: White matter hyperintensities (WMHs) are common to advancing age and likely reflect brain vascular injury among older individuals. WMH and to a lesser extent, magnetic resonance (MR) infarcts, affect risk for transition to cognitive impairment. WMHs and MR infarcts are present, even among Alzheimer's Disease Neuroimaging Initiative (ADNI) participants highly selected to have Alzheimer's disease (AD) as the primary pathology. WMH burden in ADNI is greater among individuals with cognitive impairment and has been associated with AD neurodegenerative markers and cerebral amyloidosis. The negative additive effects of cerebrovascular disease appear present, even in select populations, and future biomarker work needs to further explore this relationship.

Sex-specific topological structure associated with dementia via latent space estimation

INTRODUCTION

We investigate sex-specific topological structures associated with typical Alzheimer's disease (AD) dementia using a novel state-of-the-art latent space estimation technique.

METHODS

This study applies a probabilistic approach for latent space estimation that extends current multiplex network modeling approaches and captures the higher-order dependence in functional connectomes by preserving transitivity and modularity structures.

RESULTS

We find sex differences in network topology with females showing more default mode network (DMN)-centered hyperactivity and males showing more limbic system (LS)-centered hyperactivity, while both show DMN-centered hypoactivity. We find that centrality plays an important role in dementia-related dysfunction with stronger association between connectivity changes and regional centrality in females than in males.

DISCUSSION

The study contributes to the current literature by providing a more comprehensive picture of dementia-related neurodegeneration linking centrality, network segregation, and DMN-centered changes in functional connectomes, and how these components of neurodegeneration differ between the sexes.

HIGHLIGHTS

We find evidence supporting the active role network topology plays in neurodegeneration with an imbalance between the excitatory and inhibitory mechanisms that can lead to whole-brain destabilization in dementia patients. We find sex-based differences in network topology with females showing more default mode network (DMN)-centered hyperactivity, males showing more limbic system (LS)-centered hyperactivity, while both show DMN-centered hypoactivity. We find that brain region centrality plays an important role in dementia-related dysfunction with a stronger association between connectivity changes and regional centrality in females than in males. Females, compared to males, tend to exhibit stronger dementia-related changes in regions that are the central actors of the brain networks. Taken together, this research uniquely contributes to the current literature by providing a more comprehensive picture of dementia-related neurodegeneration linking centrality, network segregation, and DMN-centered changes in functional connectomes, and how these components of neurodegeneration differ between the sexes.

The ADNI Administrative Core: Ensuring ADNI's success and informing future AD clinical trials

The Alzheimer's Disease Neuroimaging Initiative (ADNI) Administrative Core oversees and coordinates all ADNI activities, to ensure the success and maximize the impact of ADNI in advancing Alzheimer's disease (AD) research and clinical trials. It manages finances and develops policies for data sharing, publications using ADNI data, and access to ADNI biospecimens. The Core develops and executes pilot projects to guide future ADNI activities and identifies key innovative methods for inclusion in ADNI. For ADNI4, the Administrative Core collaborates with the Engagement, Clinical, and Biomarker Cores to develop and evaluate novel, digital methods and infrastructure for participant recruitment, screening, and assessment of participants. The goal of these efforts is to enroll 500 participants, including > 50% from underrepresented populations, 40% with mild cognitive impairment, and 80% with elevated AD biomarkers. This new approach also provides a unique opportunity to validate novel methods. HIGHLIGHTS: The Alzheimer's Disease Neuroimaging Initiative (ADNI) Administrative Core oversees and coordinates all ADNI activities. The overall goal is to ensure ADNI's success and help design future Alzheimer's disease (AD) clinical trials. A key innovation is data sharing without embargo to maximize scientific impact. For ADNI4, novel, digital methods for recruitment and assessment were developed. New methods are designed to improve the participation of underrepresented populations.

Cerebral perfusion and amyloidosis in the oldest-old

INTRODUCTION

In a nested case-control study, we examined how cerebral perfusion relates to cognitive status and amyloid in the oldest-old (i.e., 90 years of age and older).

METHODS

Study participants included 113 dementia-free older adults (76 cognitively normal [CN]; 37 cognitively impaired, no dementia [CIND]) from the 90+ Study (mean age = 92.9, SD = 2.4). We quantified regional perfusion from arterial spin labeling-MRI (magnetic resonance imaging) and amyloid deposition from florbetapir-positron emission tomography (PET) in a region comprising the posterior cingulate and precuneus (PCC+PCu), and additionally quantified perfusion in other regions important for cognitive decline (medial temporal lobe, inferior parietal lobe, and orbitofrontal cortex).

RESULTS

Participants with CIND displayed lower perfusion in the PCC+PCu relative to participants who were CN, but there was no statistically significant difference between the groups in amyloid burden in this region. In addition, participants with CIND exhibited lower inferior parietal and higher orbitofrontal perfusion.

DISCUSSION

Cerebral perfusion is related to cognitive status in the oldest-old independent of amyloidosis.

HIGHLIGHTS

Cerebral perfusion and amyloid positron emission tomography (PET) were measured in older adults: 90 years of age and older. Perfusion but not amyloid differed between cognitively impaired and normal groups. Frontal and parietal regions linked to cognitive decline had altered perfusion. Perfusion is related to cognitive status in the oldest-old independent of amyloid.

Normative Neuroimaging Library: Designing a Comprehensive and Demographically Diverse Dataset of Healthy Controls to Support Traumatic Brain Injury Diagnostic and Therapeutic Development

The past decade has seen impressive advances in neuroimaging, moving from qualitative to quantitative outputs. Available techniques now allow for the inference of microscopic changes occurring in white and gray matter, along with alterations in physiology and function. These existing and emerging techniques hold the potential of providing unprecedented capabilities in achieving a diagnosis and predicting outcomes for traumatic brain injury (TBI) and a variety of other neurological diseases. To see this promise move from the research lab into clinical care, an understanding is needed of what normal data look like for all age ranges, sex, and other demographic and socioeconomic categories. Clinicians can only use the results of imaging scans to support their decision-making if they know how the results for their patient compare with a normative standard. This potential for utilizing magnetic resonance imaging (MRI) in TBI diagnosis motivated the American College of Radiology and Cohen Veterans Bioscience to create a reference database of healthy individuals with neuroimaging, demographic data, and characterization of psychological functioning and neurocognitive data that will serve as a normative resource for clinicians and researchers for development of diagnostics and therapeutics for TBI and other brain disorders. The goal of this article is to introduce the large, well-curated Normative Neuroimaging Library (NNL) to the research community. NNL consists of data collected from ∼1900 healthy participants. The highlights of NNL are (1) data are collected across a diverse population, including civilians, veterans, and active-duty service members with an age range (18-64 years) not well represented in existing datasets; (2) comprehensive structural and functional neuroimaging acquisition with state-of-the-art sequences (including structural, diffusion, and functional MRI; raw scanner data are preserved, allowing higher quality data to be derived in the future; standardized imaging acquisition protocols across sites reflect sequences and parameters often recommended for use with various neurological and psychiatric conditions, including TBI, post-traumatic stress disorder, stroke, neurodegenerative disorders, and neoplastic disease); and (3) the collection of comprehensive demographic details, medical history, and a broad structured clinical assessment, including cognition and psychological scales, relevant to multiple neurological conditions with functional sequelae. Thus, NNL provides a demographically diverse population of healthy individuals who can serve as a comparison group for brain injury study and clinical samples, providing a strong foundation for precision medicine. Use cases include the creation of imaging-derived phenotypes (IDPs), derivation of reference ranges of imaging measures, and use of IDPs as training samples for artificial intelligence-based biomarker development and for normative modeling to help identify injury-induced changes as outliers for precision diagnosis and targeted therapeutic development. On its release, NNL is poised to support the use of advanced imaging in clinician decision support tools, the validation of imaging biomarkers, and the investigation of brain-behavior anomalies, moving the field toward precision medicine.

Reductions in the white-gray functional connectome in preclinical Alzheimer's disease and their associations with amyloid and cognition

BACKGROUND

The magnitudes and patterns of alterations of the white-gray matter (WM-GM) functional connectome in preclinical Alzheimer's disease (AD), and their associations with amyloid and cognition, remain unclear.

METHODS

We compared regional WM-GM functional connectivity (FC) and network properties in subjects with preclinical AD (or AD dementia) and controls (total n = 344). Their associations with positron emission tomography AV45-measured amyloid beta (Aβ) load and modified Preclinical Alzheimer Cognitive Composite (mPACC) scores were examined.

RESULTS

Preclinical AD subjects showed lower FC in specific WM-GM pairs and reduced segregation of control, dorsal attention, and somatomotor networks. A major portion of the reduced FC and network segregations were linked to elevated Aβ. Reduced FC of one WM-GM pair correlated with impaired mPACC. AD dementia exhibited broader reductions and stronger associations.

DISCUSSION

The WM-GM functional connectome undergoes regional and systemic dysfunctions as early as in the preclinical stage, correlating with amyloid deposition and predicting cognitive impairment.

HIGHLIGHTS

Preclinical Alzheimer's disease (AD) subjects showed lower functional connectivity in specific white-gray matter (WM-GM) pairs and reduced segregations of control, dorsal attention, and somatomotor networks. A major portion of the reduced connectivity and network segregations were linked to elevated amyloid beta load. Only one WM-GM pair's reduced connectivity was linearly correlated with impaired cognitive composite scores. AD dementia showed more extensive reductions in connectivity, network integration, and segregation, with stronger associations with amyloid elevation and cognitive impairment. The WM-GM functional connectome offers a distinct perspective for understanding changes in brain functional architecture throughout the AD continuum.

Enhancing Amyloid PET Quantification: MRI-Guided Super-Resolution Using Latent Diffusion Models

Amyloid PET imaging plays a crucial role in the diagnosis and research of Alzheimer's disease (AD), allowing non-invasive detection of amyloid-β plaques in the brain. However, the low spatial resolution of PET scans limits the accurate quantification of amyloid deposition due to partial volume effects (PVE). In this study, we propose a novel approach to addressing PVE using a latent diffusion model for resolution recovery (LDM-RR) of PET imaging. We leverage a synthetic data generation pipeline to create high-resolution PET digital phantoms for model training. The proposed LDM-RR model incorporates a weighted combination of L1, L2, and MS-SSIM losses at both noise and image scales to enhance MRI-guided reconstruction. We evaluated the model's performance in improving statistical power for detecting longitudinal changes and enhancing agreement between amyloid PET measurements from different tracers. The results demonstrate that the LDM-RR approach significantly improves PET quantification accuracy, reduces inter-tracer variability, and enhances the detection of subtle changes in amyloid deposition over time. We show that deep learning has the potential to improve PET quantification in AD, effectively contributing to the early detection and monitoring of disease progression.

Constructing and exploring neuroimaging projects: a survey from clinical practice to scientific research

Over the past decades, numerous large-scale neuroimaging projects that involved the collection and release of multimodal data have been conducted globally. Distinguished initiatives such as the Human Connectome Project, UK Biobank, and Alzheimer's Disease Neuroimaging Initiative, among others, stand as remarkable international collaborations that have significantly advanced our understanding of the brain. With the advancement of big data technology, changes in healthcare models, and continuous development in biomedical research, various types of large-scale projects are being established and promoted worldwide. For project leaders, there is a need to refer to common principles in project construction and management. Users must also adhere strictly to rules and guidelines, ensuring data safety and privacy protection. Organizations must maintain data integrity, protect individual privacy, and foster stakeholders' trust. Regular updates to legislation and policies are necessary to keep pace with evolving technologies and emerging data-related challenges. CRITICAL RELEVANCE STATEMENT: By reviewing global large-scale neuroimaging projects, we have summarized the standards and norms for establishing and utilizing their data, and provided suggestions and opinions on some ethical issues, aiming to promote higher-quality neuroimaging data development. KEY POINTS: Global neuroimaging projects are increasingly advancing but still face challenges. Constructing and utilizing neuroimaging projects should follow set rules and guidelines. Effective data management and governance should be developed to support neuroimaging projects.

Mild Cognitive Impairment: Quantifying a Qualitative Disorder

Mild cognitive impairment (MCI) has been described as a transitional state between normal aging and dementia, which can be both identified and tracked over time from qualitative and/or quantitative perspectives. Each definition of MCI involves some subjective cognitive complaint, some level of objective cognitive impairment, and generally intact daily functioning. Progression to dementia is common on follow-up in MCI, but stability and reversion to normal cognition can also occur. Quantitative methods might allow health care providers to evaluate and follow the subtle declines in MCI, as well as examine possible benefits of interventions with this at-risk condition.

Tau-network mapping of domain-specific cognitive impairment in Alzheimer's disease

Fibrillar tau gradually progresses in the brain during the course of Alzheimer's disease (AD). However, the contribution of tau accumulation in a given brain region to decline in different cognitive domains and thus phenotypic heterogeneity in AD remains unclear. Here, we leveraged the functional connectome to link the locality of tau accumulation to domain-specific cognitive impairment. In the current study, we mapped regional tau-PET accumulation onto the normative functional connectome. Subsequently, we cross-validated in two samples of AD-patients the associations between the tau-connectivity profiles and cognitive domains (episodic memory, executive function, or language). Lastly, we tested the effect of local tau-PET accumulation on the domain-specific tau-lesion networks and cognition. We identified cognitive-domain-specific tau-lesion networks, where closer topological proximity of tau-PET locations to a network was predictive of worse impairment in that domain. Higher tau-PET was associated with decreased domain-specific network connectivity, and the decrease in connectivity was associated with lower domain-specific cognition. The tau locations' connectivity profile explained domain-specific cognitive impairment, where disrupted connectivity may underlie the effect of tau on cognitive impairment.

Study protocol: Cerebral autoregulation, brain perfusion, and neurocognitive outcomes after traumatic brain injury -CAPCOG-TBI

Background

Moderate-severe traumatic brain injury (msTBI) stands as a prominent etiology of adult disability, with increased risk for cognitive impairment and dementia. Although some recovery often occurs within the first year post-injury, predicting long-term cognitive outcomes remains challenging, partly due to the significant pathophysiological heterogeneity of TBI, including acute cerebrovascular injury. The primary aim of our recently funded study, cerebral autoregulation, brain perfusion, and neurocognitive outcomes after traumatic brain injury (CAPCOG-TBI), is to determine if acute cerebrovascular dysfunction after msTBI measured using multimodal non-invasive neuromonitoring is associated with cognitive outcome at 1-year post-injury.

Methods

This longitudinal observational study will be conducted at two Level 1 trauma centers in Texas, USA, and will include adult patients with msTBI, and/or mild TBI with neuroimaging abnormalities. Multimodal cerebral vascular assessment using transcranial Doppler and cerebral near-infrared spectroscopy (NIRS) will be conducted within 7-days of onset of TBI. Longitudinal outcomes, including cognitive/functional assessments (Glasgow Outcome Scale and Patient-Reported Outcomes Measurement Information System), cerebral vascular assessment, and imaging will be performed at follow-ups 3-, 6-, and 12-months post-injury. We aim to recruit 100 subjects with msTBI along with 30 orthopedic trauma controls (OTC). This study is funded by National Institute of Neurological Disease and Stroke (NINDS) and is registered on Clinicaltrial.org (NCT06480838).

Expected results

We anticipate that msTBI patients will exhibit impaired cerebrovascular function in the acute phase compared to the OTC group. The severity of cerebrovascular dysfunction during this stage is expected to inversely correlate with cognitive and functional outcomes at 1-year post-injury. Additionally, recovery from cerebrovascular dysfunction is expected to be linked to cognitive recovery.

Conclusion

The results of this study could help to understand the contribution of cerebrovascular dysfunction to cognitive outcomes after TBI and pave the way for innovative vascular-focused interventions aimed at enhancing cognitive recovery and mitigating neurodegeneration following msTB. In addition, its focus toward personalized medicine to aid in the management and prognosis of TBI patients.

Contributions of the ADNI Biostatistics Core

The goal of the Biostatistics Core of the Alzheimer's Disease Neuroimaging Initiative (ADNI) has been to ensure that sound study designs and statistical methods are used to meet the overall goals of ADNI. We have supported the creation of a well-validated and well-curated longitudinal database of clinical and biomarker information on ADNI participants and helped to make this accessible and usable for researchers. We have developed a statistical methodology for characterizing the trajectories of clinical and biomarker change for ADNI participants across the spectrum from cognitively normal to dementia, including multivariate patterns and evidence for heterogeneity in cognitive aging. We have applied these methods and adapted them to improve clinical trial design. ADNI-4 will offer us a chance to help extend these efforts to a more diverse cohort with an even richer panel of biomarker data to support better knowledge of and treatment for Alzheimer's disease and related dementias. HIGHLIGHTS: The Alzheimer's Disease Neuroimaging Initiative (ADNI) Biostatistics Core provides study design and analytic support to ADNI investigators. Core members develop and apply novel statistical methodology to work with ADNI data and support clinical trial design. The Core contributes to the standardization, validation, and harmonization of biomarker data. The Core serves as a resource to the wider research community to address questions related to the data and study as a whole.

The Alzheimer's Disease Neuroimaging Initiative Clinical Core

The Alzheimer's Disease Neuroimaging Initiative (ADNI) Clinical Core is responsible for coordination of all clinical activities at the ADNI sites, including project management, regulatory oversight, and site management and monitoring, as well as the collection of all clinical data and management of all study data. The Clinical Core is also charged with determining the clinical classifications and criteria for enrollment in evolving AD trials and enabling the ongoing characterization of the cross-sectional features and longitudinal trajectories of the ADNI cohorts with application of these findings to optimal clinical trial designs. More than 2400 individuals have been enrolled in the cohorts since the inception of ADNI, facilitating refinement of our understanding of the AD trajectory and allowing academic and industry investigators to model therapeutic trials across the disease spectrum from the presymptomatic stage through dementia. HIGHLIGHTS: Since 2004, the Alzheimer's Disease Neuroimaging Initiative (ADNI) Clinical Core has overseen the enrollment of > 2400 participants with mild cognitive impairment, mild Alzheimer's disease (AD) dementia, and normal cognition. The longitudinal dataset has elucidated the full cognitive and clinical trajectory of AD from its presymptomatic stage through the onset of dementia. The ADNI data have supported the design of most major trials in the field.

Efficient segmentation of fetal brain MRI based on the physical resolution

BACKGROUND

The image resolution of fetal brain magnetic resonance imaging (MRI) is a critical factor in brain development measures, which is mainly determined by the physical resolution configured in the MRI sequence. However, fetal brain MRI are commonly reconstructed to 3D images with a higher apparent resolution, compared to the original physical resolution.

PURPOSE

This work is to demonstrate that accurate segmentation can be achieved based on the MRI physical resolution, and the high apparent resolution segmentation can be achieved by a simple deep learning module.

METHODS

This retrospective study included 150 adult and 80 fetal brain MRIs. The adult brain MRIs were acquired at a high physical resolution, which were downsampled to visualize and quantify its impacts on the segmentation accuracy. The physical resolution of fetal images was estimated based on MRI acquisition settings and the images were downsampled accordingly before segmentation and restored using multiple upsampling strategies. Segmentation accuracy of ConvNet models were evaluated on the original and downsampled images. Dice coefficients were calculated, and compared to the original data.

RESULTS

When the apparent resolution was higher than the physical resolution, the accuracy of fetal brain segmentation had negligible degradation (accuracy reduced by 0.26%, 1.1%, and 1.8% with downsampling factors of 4/3, 2, and 4 in each dimension, without significant differences from the original data). Using a downsampling factor of 4 in each dimension, the proposed method provided 7× smaller and 10× faster models.

CONCLUSION

Efficient and accurate fetal brain segmentation models can be developed based on the physical resolution of MRI acquisitions.

Utility of cerebrovascular imaging biomarkers to detect cerebral amyloidosis

INTRODUCTION

The relationship between cerebrovascular disease (CVD) and amyloid beta (Aβ) in Alzheimer's disease (AD) is understudied. We hypothesized that magnetic resonance imaging (MRI)-based CVD biomarkers-including cerebral microbleeds (CMBs), lacunar infarction, and white matter hyperintensities (WMHs)-would correlate with Aβ positivity on positron emission tomography (Aβ-PET).

METHODS

We cross-sectionally analyzed data from the Alzheimer's Disease Neuroimaging Initiative (ADNI, N = 1352). Logistic regression was used to calculate odds ratios (ORs), with Aβ-PET positivity as the standard-of-truth.

RESULTS

Following adjustment, WMHs (OR = 1.25) and superficial CMBs (OR = 1.45) remained positively associated with Aβ-PET positivity (p < 0.001). Deep CMBs and lacunes exhibited a varied relationship with Aβ-PET in cognitive subgroups. The combined diagnostic model, which included CVD biomarkers and other accessible measures, significantly predicted Aβ-PET (pseudo-R2 = 0.41).

DISCUSSION

The study highlights the translational value of CVD biomarkers in diagnosing AD, and underscores the need for more research on their inclusion in diagnostic criteria.

CLINICALTRIALS

gov: ADNI-2 (NCT01231971), ADNI-3 (NCT02854033).

HIGHLIGHTS

Cerebrovascular biomarkers linked to amyloid beta (Aβ) in Alzheimer's disease (AD). White matter hyperintensities and cerebral microbleeds reliably predict Aβ-PET positivity. Relationships with Aβ-PET vary by cognitive stage. Novel accessible model predicts Aβ-PET status. Study supports multimodal diagnostic approaches.

Predicting CDR status over 36 months with a recall-based digital cognitive biomarker

INTRODUCTION

Word-list recall tests are routinely used for cognitive assessment, and process scoring may improve their accuracy. We examined whether Alzheimer's Disease Assessment Scale-Cognitive subscale (ADAS-Cog) derived, process-based digital cognitive biomarkers (DCBs) at baseline predicted Clinical Dementia Rating (CDR) longitudinally and compared them to standard metrics.

METHODS

Analyses were performed with Alzheimer's Disease Neuroimaging Initiative (ADNI) data from 330 participants (mean age = 71.4 ± 7.2). We conducted regression analyses predicting CDR at 36 months, controlling for demographics and genetic risk, with ADAS-Cog traditional scores and DCBs as predictors.

RESULTS

The best predictor of CDR at 36 months was M, a DCB reflecting recall ability (area under the curve = 0.84), outperforming traditional scores. Diagnostic results suggest that M may be particularly useful to identify individuals who are unlikely to decline.

DISCUSSION

These results suggest that M outperforms ADAS-Cog traditional metrics and supports process scoring for word-list recall tests. More research is needed to determine further applicability with other tests and populations.

HIGHLIGHTS

Process scoring and latent modeling were more effective than traditional scoring. Latent recall ability (M) was the best predictor of Clinical Dementia Rating decline at 36 months. The top digital cognitive biomarker model had odds ≈ 90 times greater than the top Alzheimer's Disease Assessment Scale-Cognitive subscale model. Particularly high negative predictive value supports literature on cognitive testing as a useful screen. Consideration of both cognitive and pathological outcomes is needed.

Fingerprints of brain disease: connectome identifiability in Alzheimer's disease

Functional connectivity patterns in the human brain, like the friction ridges of a fingerprint, can uniquely identify individuals. Does this "brain fingerprint" remain distinct even during Alzheimer's disease (AD)? Using fMRI data from healthy and pathologically ageing subjects, we find that individual functional connectivity profiles remain unique and highly heterogeneous during mild cognitive impairment and AD. However, the patterns that make individuals identifiable change with disease progression, revealing a reconfiguration of the brain fingerprint. Notably, connectivity shifts towards functional system connections in AD and lower-order cognitive functions in early disease stages. These findings emphasize the importance of focusing on individual variability rather than group differences in AD studies. Individual functional connectomes could be instrumental in creating personalized models of AD progression, predicting disease course, and optimizing treatments, paving the way for personalized medicine in AD management.

Fixel-based and tensor-derived white matter abnormalities in relation to memory impairment and neurocognitive disorders

Aging-related neurocognitive disorders, including Alzheimer's disease (AD) and mild cognitive impairment (MCI), have been characterised by altered brain white matter (WM), relying widely on diffusion tensor imaging (DTI). DTI's limited accuracy in assessing crossing fibres prompted novel methods that distinguish fibres crossing through same voxel-spaces, such as fixel-based analysis (FBA), highlighting subtle macrostructural and microstructural alterations in AD and MCI. We examined the FBA and DTI's specificity in determining WM features relevant to memory in the neurocognitive aging spectrum. Diffusion-weighted images were analysed in 560 participants with various neurocognitive diagnoses from the Alzheimer's Disease Neuroimaging Initiative (F:297; mean age: 73.2 ± 8). Verbal memory was measured in 488 participants using the Rey Auditory Verbal Learning Test. FBA-derived measures of fibre density (FD), fibre-bundle cross-section (FC), and their combination (FDC), DTI fractional anisotropy (FA) and mean diffusivity (MD) were examined in relation to diagnoses and memory scores, controlling for age, sex, and intracranial volume. MCI and AD groups significantly differed from controls, with lower FD and FDC in the fornix and bilateral fibres extending to the medial temporal lobes (MTL). Memory was positively associated with FD and FDC in the fornix and MTL fibres, and FC in the anterior commissure (AC). Widespread FA reductions and MD increases were observed in AD and MCI and widely associated with memory. Fixel-wise measures highlight fibre tracts that are altered distinctly at the macroscopic and microscopic level in neurocognitive aging, and reveals structures associated with memory performance that are more specifically located than tensor-derived measures.

Progressive verbal apraxia of reading

We identified a syndrome characterized by a relatively isolated progressive impairment of reading words that the patient was able to understand and repeat but without other components of speech apraxia. This cluster of symptoms fits a new syndrome designated Progressive Verbal Apraxia of Reading. A right-handed man (AB) came with a 2.5-year history of increasing difficulties in reading aloud. He was evaluated twice, 2 years apart, using multimodal neuroimaging techniques and quantitative neurolinguistic assessment. In the laboratory, reading difficulties arose in the context of intact visual and auditory word recognition as well as intact ability to understand and repeat words he was unable to read aloud. The unique feature was the absence of dysarthria or speech apraxia in tasks other than reading. Initial imaging did not reveal statistically significant atrophy. Structural magnetic resonance and FDG-PET imaging at the second assessment revealed atrophy and hypometabolism in the right posterior cerebellum, in areas shown to be part of his language network by task-based functional neuroimaging at initial assessment. This syndromic cluster can be designated Progressive Verbal Apraxia of Reading, an entity that has not been reported previously to the best of our knowledge. We hypothesize a selective disconnection of the visual word recognition system from the otherwise intact articulatory apparatus, a disconnection that appears to reflect the disruption of multisynaptic cerebello-cortical circuits.

Deep learning for the harmonization of structural MRI scans: a survey

Medical imaging datasets for research are frequently collected from multiple imaging centers using different scanners, protocols, and settings. These variations affect data consistency and compatibility across different sources. Image harmonization is a critical step to mitigate the effects of factors like inherent differences between various vendors, hardware upgrades, protocol changes, and scanner calibration drift, as well as to ensure consistent data for medical image processing techniques. Given the critical importance and widespread relevance of this issue, a vast array of image harmonization methodologies have emerged, with deep learning-based approaches driving substantial advancements in recent times. The goal of this review paper is to examine the latest deep learning techniques employed for image harmonization by analyzing cutting-edge architectural approaches in the field of medical image harmonization, evaluating both their strengths and limitations. This paper begins by providing a comprehensive fundamental overview of image harmonization strategies, covering three critical aspects: established imaging datasets, commonly used evaluation metrics, and characteristics of different scanners. Subsequently, this paper analyzes recent structural MRI (Magnetic Resonance Imaging) harmonization techniques based on network architecture, network learning algorithm, network supervision strategy, and network output. The underlying architectures include U-Net, Generative Adversarial Networks (GANs), Variational Autoencoders (VAEs), flow-based generative models, transformer-based approaches, as well as custom-designed network architectures. This paper investigates the effectiveness of Disentangled Representation Learning (DRL) as a pivotal learning algorithm in harmonization. Lastly, the review highlights the primary limitations in harmonization techniques, specifically the lack of comprehensive quantitative comparisons across different methods. The overall aim of this review is to serve as a guide for researchers and practitioners to select appropriate architectures based on their specific conditions and requirements. It also aims to foster discussions around ongoing challenges in the field and shed light on promising future research directions with the potential for significant advancements.

BundleAGE: Predicting White Matter Age using Along-Tract Microstructural Profiles from Diffusion MRI

Brain Age Gap Estimation (BrainAGE) is an estimate of the gap between a person's chronological age (CA) and a measure of their brain's 'biological age' (BA). This metric is often used as a marker of accelerated aging, albeit with some caveats. Age prediction models trained on brain structural and functional MRI have been employed to derive BrainAGE biomarkers, for predicting the risk of neurodegeneration. While voxel-based and along-tract microstructural maps from diffusion MRI have been used to study brain aging, no studies have evaluated along-tract microstructure for computing BrainAGE. In this study, we train machine learning models to predict a person's age using along-tract microstructural profiles from diffusion tensor imaging. We were able to demonstrate differential aging patterns across different white matter bundles and microstructural measures. The novel Bundle Age Gap Estimation (BundleAGE) biomarker shows potential in quantifying risk factors for neurodegenerative diseases and aging, while incorporating finer scale information throughout white matter bundles.

Mapping brain morphology to cognitive deficits: a study on PD-CRS scores in Parkinson's disease with mild cognitive impairment

Background

The Parkinson's Disease-Cognitive Rating Scale (PD-CRS) is a widely used tool for detecting mild cognitive impairment (MCI) in Parkinson's Disease (PD) patients, however, the neuroanatomical underpinnings of this test's outcomes require clarification. This study aims to: (a) investigate cortical volume (CVol) and cortical thickness (CTh) disparities between PD patients exhibiting mild cognitive impairment (PD-MCI) and those with preserved cognitive abilities (PD-IC); and (b) identify the structural correlates in magnetic resonance imaging (MRI) of overall PD-CRS performance, including its subtest scores, within a non-demented PD cohort.

Materials and methods

This study involved 51 PD patients with Hoehn & Yahr stages I-II, categorized into two groups: PD-IC (n = 36) and PD-MCI (n = 15). Cognitive screening evaluations utilized the PD-CRS and the Montreal Cognitive Assessment (MoCA). PD-MCI classification adhered to the Movement Disorder Society Task Force criteria, incorporating extensive neuropsychological assessments. The interrelation between brain morphology and cognitive performance was determined using FreeSurfer.

Results

Vertex-wise analysis of the entire brain demonstrated a notable reduction in CVol within a 2,934 mm2 cluster, encompassing parietal and temporal regions, in the PD-MCI group relative to the PD-IC group. Lower PD-CRS total scores correlated with decreased CVol in the middle frontal, superior temporal, inferior parietal, and cingulate cortices. The PD-CRS subtests for Sustained Attention and Clock Drawing were associated with cortical thinning in distinct regions: the Clock Drawing subtest correlated with changes in the parietal lobe, insula, and superior temporal cortex morphology; while the PD-CRS frontal-subcortical scores presented positive correlations with CTh in the transverse temporal, medial orbitofrontal, superior temporal, precuneus, fusiform, and supramarginal regions. Additionally, PD-CRS subtests for Semantic and Alternating verbal fluency were linked to CTh changes in orbitofrontal, temporal, fusiform, insula, and precentral regions.

Conclusion

PD-CRS performance mirrors neuroanatomical changes across extensive fronto-temporo-parietal areas, covering both lateral and medial cortical surfaces, in PD patients without dementia. The observed changes in CVol and CTh associated with this cognitive screening tool suggest their potential as surrogate markers for cognitive decline in PD. These findings warrant further exploration and validation in multicenter studies involving independent patient cohorts.

Mapping Knowledge Landscapes and Emerging Trends in AI for Dementia Biomarkers: Bibliometric and Visualization Analysis

BACKGROUND

With the rise of artificial intelligence (AI) in the field of dementia biomarker research, exploring its current developmental trends and research focuses has become increasingly important. This study, using literature data mining, analyzes and assesses the key contributions and development scale of AI in dementia biomarker research.

OBJECTIVE

The aim of this study was to comprehensively evaluate the current state, hot topics, and future trends of AI in dementia biomarker research globally.

METHODS

This study thoroughly analyzed the literature in the application of AI to dementia biomarkers across various dimensions, such as publication volume, authors, institutions, journals, and countries, based on the Web of Science Core Collection. In addition, scales, trends, and potential connections between AI and biomarkers were extracted and deeply analyzed through multiple expert panels.

RESULTS

To date, the field includes 1070 publications across 362 journals, involving 74 countries and 1793 major research institutions, with a total of 6455 researchers. Notably, 69.41% (994/1432) of the researchers ceased their studies before 2019. The most prevalent algorithms used are support vector machines, random forests, and neural networks. Current research frequently focuses on biomarkers such as imaging biomarkers, cerebrospinal fluid biomarkers, genetic biomarkers, and blood biomarkers. Recent advances have highlighted significant discoveries in biomarkers related to imaging, genetics, and blood, with growth in studies on digital and ophthalmic biomarkers.

CONCLUSIONS

The field is currently in a phase of stable development, receiving widespread attention from numerous countries, institutions, and researchers worldwide. Despite this, stable clusters of collaborative research have yet to be established, and there is a pressing need to enhance interdisciplinary collaboration. Algorithm development has shown prominence, especially the application of support vector machines and neural networks in imaging studies. Looking forward, newly discovered biomarkers are expected to undergo further validation, and new types, such as digital biomarkers, will garner increased research interest and attention.