The Australian Imaging, Biomarkers and Lifestyle (AIBL) study of aging: methodology and baseline characteristics of 1112 individuals recruited for a longitudinal study of Alzheimer's disease

A computational ontology framework for the synthesis of multi-level pathology reports from brain MRI scans

BackgroundConvolutional neural network (CNN) based volumetry of MRI data can help differentiate Alzheimer's disease (AD) and the behavioral variant of frontotemporal dementia (bvFTD) as causes of cognitive decline and dementia. However, existing CNN-based MRI volumetry tools lack a structured hierarchical representation of brain anatomy, which would allow for aggregating regional pathological information and automated computational inference.ObjectiveDevelop a computational ontology pipeline for quantifying hierarchical pathological abnormalities and visualize summary charts for brain atrophy findings, aiding differential diagnosis.MethodsUsing FastSurfer, we segmented brain regions and measured volume and cortical thickness from MRI scans pooled across multiple cohorts (N = 3433; ADNI, AIBL, DELCODE, DESCRIBE, EDSD, and NIFD), including healthy controls, prodromal and clinical AD cases, and bvFTD cases. Employing the Web Ontology Language (OWL), we built a semantic model encoding hierarchical anatomical information. Additionally, we created summary visualizations based on sunburst plots for visual inspection of the information stored in the ontology.ResultsOur computational framework dynamically estimated and aggregated regional pathological deviations across different levels of neuroanatomy abstraction. The disease similarity index derived from the volumetric and cortical thickness deviations achieved an AUC of 0.88 for separating AD and bvFTD, which was also reflected by distinct atrophy profile visualizations.ConclusionsThe proposed automated pipeline facilitates visual comparison of atrophy profiles across various disease types and stages. It provides a generalizable computational framework for summarizing pathologic findings, potentially enhancing the physicians' ability to evaluate brain pathologies robustly and interpretably.

MRI Distance Measures as a Predictor of Subsequent Clinical Status During the Preclinical Phase of Alzheimer's Disease and Related Disorders

Brain atrophy over time, as measured by magnetic resonance imaging (MRI), has been shown to predict subsequent cognitive impairment among individuals who were cognitively normal when first evaluated, indicating that subtle brain atrophy associated with Alzheimer's disease (AD) may begin years before clinical symptoms appear. Traditionally, atrophy has been quantified by differences in brain volume or thickness over a specified timeframe. Research indicates that the rate of atrophy varies across different brain regions, which themselves exhibit complex spatial and hierarchical organizations. These characteristics collectively emphasize the need for diverse summary measures that can effectively capture the multidimensional nature of degeneration. In this study, we explore the use of distance measurements to quantify brain volumetric changes using processed MRI data from the Preclinical Alzheimer's Disease Consortium (PAC). We conducted a series of analyses to predict future diagnostic status by modeling MRI trajectories for participants who were cognitively normal at baseline and either remained cognitively normal or progressed to mild cognitive impairment (MCI) over time, with adjustments for age, sex, education, and APOE genotype. We consider multiple distance measures and brain regions through a two-step approach. First, we build base models by fitting individual mixed-effect models for each distance metric and brain region pairing, using follow-up diagnosis (normal vs. MCI) as the outcome and volumetric changes from the baseline, as summarized by a given distance measure, as predictors. The second step aggregates these individual region-distance base models to derive an overall estimate of diagnostic status. Our analyses showed that the distance measures approach consistently outperformed the traditional direct volumetric approach in terms of predictive accuracy, both in individual base models and the aggregated models. This work highlights the potential advantage of using distance measures over the traditional direct volumetric approach to capture the multidimensional aspects of atrophy in the development of AD and related disorders.

Baseline habitual dietary nitrate intake and Alzheimer's Disease related neuroimaging biomarkers in the Australian Imaging, Biomarkers and Lifestyle study of ageing

BACKGROUND

Dietary nitrate, as a nitric oxide (NO) precursor, may support brain health and protect against dementia.

OBJECTIVE

Our primary aim was to investigate whether dietary nitrate is associated with neuroimaging markers of brain health linked with Alzheimer's disease (AD).

PARTICIPANTS

Study participants were cognitively unimpaired individuals from the Australian Imaging, Biomarkers and Lifestyle Study of Ageing (AIBL) who had β-amyloid positron emission tomography (PET) scans (n = 554) and magnetic resonance imaging (MRI) scans (n = 335) and had completed a Food Frequency Questionnaire at baseline.

METHODS

Source-specific nitrate intakes were estimated using comprehensive nitrate food composition databases. Rates of cerebral β-amyloid (Aβ) deposition, measured using PET, and rates of brain atrophy, measured using MRI, were assessed between baseline and 126-months follow-up, at intervals of 18 months. Multivariable-adjusted linear mixed effect models were used to examine associations between baseline source-specific nitrate intake and rates of (i) cerebral Aβ deposition and (ii) brain atrophy, over the 126 months of follow-up. Analyses were carried out following stratification of the sample by established dementia Alzheimer's disease (AD) risk factors including sex and presence or absence of the apolipoprotein E (APOE) ε4 allele.

RESULTS

In women carriers of the APOE ε4 allele, higher plant sourced nitrate intake (median intake 121 mg/day), was associated with a slower rate of cerebral Aβ deposition [β: 4.47 versus 8.99 Centiloid (CL) /18 months, p < 0.05] and right hippocampal atrophy [-0.01 versus -0.03 mm3 /18 months, p < 0.01], after multivariable adjustments. Moderate intake showed protective associations in men carriers and in both men and women non-carriers of APOE ε4.

CONCLUSIONS

Associations were observed between plant-derived nitrate intake and cerebral Aβ deposition, particularly in high-risk populations (women and APOE ε4 carriers). Associations were also observed for brain volume atrophy, however these exhibited subgroup variability without clear patterns relative to sex and APOE ε4 allele carriage. These findings suggest a potential link between plant-sourced nitrate and AD related neuroimaging markers of brain health improved brain health, but further validation in larger studies is required.

A lightweight generative model for interpretable subject-level prediction

Recent years have seen a growing interest in methods for predicting an unknown variable of interest, such as a subject's diagnosis, from medical images depicting its anatomical-functional effects. Methods based on discriminative modeling excel at making accurate predictions, but are challenged in their ability to explain their decisions in anatomically meaningful terms. In this paper, we propose a simple technique for single-subject prediction that is inherently interpretable. It augments the generative models used in classical human brain mapping techniques, in which the underlying cause-effect relations can be encoded, with a multivariate noise model that captures dominant spatial correlations. Experiments demonstrate that the resulting model can be efficiently inverted to make accurate subject-level predictions, while at the same time offering intuitive visual explanations of its inner workings. The method is easy to use: training is fast for typical training set sizes, and only a single hyperparameter needs to be set by the user. Our code is available at https://github.com/chiara-mauri/Interpretable-subject-level-prediction.

Pioneering new paths: the role of generative modelling in neurological disease research

Recently, deep generative modelling has become an increasingly powerful tool with seminal work in a myriad of disciplines. This powerful modelling approach is supposed to not only have the potential to solve current problems in the medical field but also to enable personalised precision medicine and revolutionise healthcare through applications such as digital twins of patients. Here, the core concepts of generative modelling and popular modelling approaches are first introduced to consider the potential based on methodological concepts for the generation of synthetic data and the ability to learn a representation of observed data. These potentials will be reviewed using current applications in neuroimaging for data synthesis and disease decomposition in Alzheimer's disease and multiple sclerosis. Finally, challenges for further research and applications will be discussed, including computational and data requirements, model evaluation, and potential privacy risks.

AI-driven fusion of neurological work-up for assessment of biological Alzheimer's disease

Alzheimer's disease (AD) diagnosis hinges on detecting amyloid beta (Aβ) plaques and neurofibrillary tau (τ) tangles. While amyloid PET imaging is now clinically approved, tau PET remains largely restricted to research settings. These imaging techniques, though valuable, are expensive and often difficult to access, limiting their widespread use in routine clinical practice. Here, we introduce a computational framework that leverages multimodal data from seven distinct cohorts comprising 12, 185 participants to estimate individual PET profiles, both global and regional, using more accessible data modalities, such as demographics, medical history, medication use, fluid measurements, functional and neuropsychological assessments, and structural MRIs. Our approach achieved an area under the receiver operating characteristic curve of 0.79 and 0.84 in classifying persons with positive Aβ and τ status, respectively. Model predictions were consistent with various biomarker and cognitive profiles, as well as with different degrees of protein abnormalities observed in post-mortem examinations. Furthermore, the regional volumes identified by the model as important aligned with the spatial distributions of the standardized uptake value ratio for regional τ labels. Our model offers a practical approach to identify potential candidates for newly approved anti-amyloid treatments and AD clinical trials for combined amyloid and tau therapies by utilizing standard neurological evaluation data.

Sex Differences in Apolipoprotein E and Alzheimer Disease Pathology Across Ancestries

Importance

Age, sex, and apolipoprotein E (APOE) are the strongest risk factors for late-onset Alzheimer disease (AD). The role of APOE in AD varies with sex and ancestry. While the association of APOE with AD biomarkers also varies across sex and ancestry, no study has systematically investigated both sex-specific and ancestry differences of APOE on cerebrospinal fluid (CSF) biomarkers together, resulting in limited insights and generalizability.

Objective

To systematically investigate the association of sex and APOE-ε4 with 3 core CSF biomarkers across ancestries.

Design, Setting, and Participants

This cohort study examined 3 CSF biomarkers (amyloid β1-42 [Aβ42], phosphorylated tau 181 [p-tau], and total tau, in participants from 20 cohorts from July 1, 1985, to March 31, 2020. These individuals were grouped into African, Asian, and European ancestries based on genetic data. Data analyses were conducted from June 1, 2023, to November 10, 2024.

Exposure

Sex (male or female) and APOE-ε4.

Main Outcomes and Measures

The associations of sex and APOE-ε4 with biomarker levels were assessed within each ancestry group, adjusting for age. Meta-analyses were performed to identify these associations across ancestries. Sensitivity analyses were conducted to exclude the potential influence of the APOE-ε2 allele.

Results

This cohort study included 4592 individuals (mean [SD] age, 70.8 [10.2] years; 2425 [52.8%] female; 119 [2.6%] African, 52 [1.1%] Asian, and 4421 [96.3%] European). Higher APOE-ε4 dosage scores were associated with lower Aβ42 values (β [SE], -0.58 [0.02], P < .001), indicating more severe pathology; these associations were seen in men and women separately and jointly. The association with APOE-ε4 was statistically greater in men (β [SE], -0.63 [0.03]; P < .001) vs women (β [SE], -0.52 [0.03]; P < .001) of European ancestry (P = .01 for interaction). Women had higher levels of p-tau, indicating more severe neurofibrillary pathology. The association between APOE-ε4 dosage and p-tau was in the expected direction (higher APOE-ε4 dosage for higher p-tau values) in both sexes, but the difference between sexes was significant only in those of African ancestry (β [SE], 0.10 [0.18]; P = .57 for men; β [SE], 0.66 [0.17]; P < .001 for women; P = .03 for interaction). Women also had higher levels of total tau, indicating more neuronal damage. The association between APOE-ε4 dosage and total tau was stronger in women than in men in the African cohort (β [SE], 0.20 [0.22]; P = .36 for men and β [SE], 0.65 [0.22], P = .004 for women [P = .16 for interaction]) and European cohort (β [SE], 0.36 [0.03]; P < .001 in women and β [SE], 0.27 [0.03], P < .001 in men [P = .053 for interaction]); no significant associations were found in the Asian cohort. Sensitivity analysis excluding APOE-ε2 carriers yielded similar results.

Conclusions and Relevance

In this cohort study, the association of the APOE-ε4 risk allele with tau accumulation was higher in women than in men. These findings underscore the importance of considering sex differences in APOE-ε4's association with AD biomarkers and tau pathology mechanisms in AD. Although this study provides robust evidence of complex interplay between sex and APOE-ε4 for European ancestry, further research is needed to fully understand other ancestry differences.

A mini review of transforming dementia care in China with data-driven insights: overcoming diagnostic and time-delayed barriers

Introduction

Inadequate primary care infrastructure and training in China and misconceptions about aging lead to high mis-/under-diagnoses and serious time delays for dementia patients, imposing significant burdens on family members and medical carers.

Main body

A flowchart integrating rural and urban areas of China dementia care pathway is proposed, especially spotting the obstacles of mis/under-diagnoses and time delays that can be alleviated by data-driven computational strategies. Artificial intelligence (AI) and machine learning models built on dementia data are succinctly reviewed in terms of the roadmap of dementia care from home, community to hospital settings. Challenges and corresponding recommendations to clinical transformation are then reported from the viewpoint of diverse dementia data integrity and accessibility, as well as models' interpretability, reliability, and transparency.

Discussion

Dementia cohort study along with developing a center-crossed dementia data platform in China should be strongly encouraged, also data should be publicly accessible where appropriate. Only be doing so can the challenges be overcome and can AI-enabled dementia research be enhanced, leading to an optimized pathway of dementia care in China. Future policy-guided cooperation between researchers and multi-stakeholders are urgently called for dementia 4E (early-screening, early-assessment, early-diagnosis, and early-intervention).

The moderating effect of diet on the relationship between depressive symptoms and Alzheimer's disease-related blood-based biomarkers

Associations between mental health, diet, and risk of Alzheimer's disease highlight the need to investigate whether dietary patterns moderate the relationship between symptoms of depression and anxiety, and neurodegeneration-related blood-based biomarkers. Cognitively unimpaired participants (n = 89) were included from the Australian Imaging, Biomarkers and Lifestyle study (mean age 75.37; 44 % male). Participants provided dietary, depressive and anxiety symptom data, and had measurement of blood-based biomarkers. Dietary pattern scores (Mediterranean diet (MeDi), Dietary Approaches to Stop Hypertension diet (DASH), and Western diet) were generated. Moderation and simple slope analyses were employed. In males with mean and below mean MeDi adherence, depressive symptoms were associated with higher neurofilament light (NfL) levels. In Apolipoprotein E ε4 non-carriers with lower than mean and mean MeDi adherence, depressive symptoms were associated with higher NfL and Aβ40 levels. No associations were observed between DASH and Western diets and neurodegeneration-related biomarkers. MeDi adherence is potentially a moderator of the relationship between depressive symptoms and neurodegeneration-related blood-based biomarkers, with sex- and genotype-specific approaches important to consider within this relationship.

Predicting amyloid beta accumulation in cognitively unimpaired older adults: Cognitive assessments provide no additional utility beyond demographic and genetic factors

BACKGROUND

Integrating non-invasive measures to estimate abnormal amyloid beta accumulation (Aβ+) is key to developing a screening tool for preclinical Alzheimer's disease (AD). The predictive capability of standard neuropsychological tests in estimating Aβ+ has not been quantified.

METHODS

We constructed machine learning models using six cognitive measurements alongside demographic and genetic risk factors to predict Aβ status. Data were drawn from three cohorts: Anti-Amyloid Treatment in Asymptomatic Alzheimer's Disease (A4), Alzheimer's Disease Neuroimaging Initiative (ADNI), and Australian Imaging, Biomarker & Lifestyle (AIBL) study. Internal validation was conducted within A4 with external validations in ADNI and AIBL to assess model generalizability.

RESULTS

The highest area under the curve (AUC) for predicting Aβ+ was observed with demographic, genetic, and cognitive variables in A4 (median AUC = 0.745), but this was not significantly different from models without cognitive variables. External validation showed no improvement in ADNI and a slight decrease in AIBL.

DISCUSSION

Standard neuropsychological tests do not significantly enhance Aβ+ prediction in cognitively unimpaired adults beyond demographic and genetic information.

HIGHLIGHTS

Standard neuropsychological tests do not significantly improve the prediction of amyloid beta positivity (Aβ+) in cognitively unimpaired older adults beyond demographic and genetic information alone. Across three well-characterized cohorts, machine learning models incorporating cognitive measures failed to significantly improve Aβ+ prediction, indicating the limited relationship between cognitive performance on these tests and the risk of pre-clinical Alzheimer's disease (AD). These findings challenge assumptions about cognitive symptoms preceding Aβ+ screening and emphasize the need for developing more sensitive cognitive tests for early AD detection.

Subtypes of brain change in aging and their associations with cognition and Alzheimer's disease biomarkers

Structural brain changes underlie cognitive changes and interindividual variability in cognition in older age. By using structural MRI data-driven clustering, we aimed to identify subgroups of cognitively unimpaired older adults based on brain change patterns and assess how changes in cortical thickness, surface area, and subcortical volume relate to cognitive change. We tested (1) which brain structural changes predict cognitive change (2) whether these are associated with core cerebrospinal fluid (CSF) Alzheimer's disease biomarkers, and (3) the degree of overlap between clusters derived from different structural modalities in 1899 cognitively healthy older adults followed up to 16 years. We identified four groups for each brain feature, based on the degree of a main longitudinal component of decline. The minimal overlap between features suggested that each contributed uniquely and independently to structural brain changes in aging. Cognitive change and baseline cognition were associated with cortical area change, whereas higher baseline levels of phosphorylated tau and amyloid-β related to changes in subcortical volume. These results may contribute to a better understanding of different aging trajectories.

Neurofind: using deep learning to make individualised inferences in brain-based disorders

Within precision psychiatry, there is a growing interest in normative models given their ability to parse heterogeneity. While they are intuitive and informative, the technical expertise and resources required to develop normative models may not be accessible to most researchers. Here we present Neurofind, a new freely available tool that bridges this gap by wrapping sound and previously tested methods on data harmonisation and advanced normative models into a web-based platform that requires minimal input from the user. We explain how Neurofind was developed, how to use the Neurofind website in four simple steps ( www.neurofind.ai ), and provide exemplar applications. Neurofind takes as input structural MRI images and outputs two main metrics derived from independent normative models: (1) Outlier Index Score, a deviation score from the normative brain morphology, and (2) Brain Age, the predicted age based on an individual's brain morphometry. The tool was trained on 3362 images of healthy controls aged 20-80 from publicly available datasets. The volume of 101 cortical and subcortical regions was extracted and modelled with an adversarial autoencoder for the Outlier index model and a support vector regression for the Brain age model. To illustrate potential applications, we applied Neurofind to 364 images from three independent datasets of patients diagnosed with Alzheimer's disease and schizophrenia. In Alzheimer's disease, 55.2% of patients had very extreme Outlier Index Scores, mostly driven by larger deviations in temporal-limbic structures and ventricles. Patients were also homogeneous in how they deviated from the norm. Conversely, only 30.1% of schizophrenia patients were extreme outliers, due to deviations in the hippocampus and pallidum, and patients tended to be more heterogeneous than controls. Both groups showed signs of accelerated brain ageing.

SCD-plus features and AD biomarkers in cognitively unimpaired samples: A meta-analytic approach for nine cohort studies

INTRODUCTION

Specific features of subjective cognitive decline (SCD-plus) have been proposed to indicate an increased risk of preclinical Alzheimer's disease (AD). However, few studies have examined how these features relate to AD biomarkers in cognitively unimpaired (CU) older adults.

METHODS

Meta-analyses were performed using cross-sectional data from nine cohorts (n = 7219, mean age (SD): 71.17 (5.9), 56.5% female) to determine associations of SCD-plus features with positron emission tomography (PET)- or cerebrospinal fluid (CSF)-derived amyloid beta (Aβ) and tau biomarkers.

RESULTS

Participants with preclinical AD (community-based only) were more likely to fulfill SCD-plus features. The presence of self-reported memory decline, associated concern/worry, and a higher number of fulfilled features were all associated with high Aβ levels. Only the latter was associated with abnormal tau.

DISCUSSION

Simultaneous endorsement of multiple SCD-plus features is a robust indicator of abnormal AD biomarkers in CU older adults, whereas isolated SCD features seem only sensitive to elevated Aβ, supporting their value as early behavioral markers of preclinical AD.

HIGHLIGHTS

About two-tenths of our sample had abnormal amyloid beta (Aβ) levels with evidence of subjective cognitive decline (SCD). Preclinical AD subsamples (community-based) had a higher percentage of participants meeting SCD-plus features. Self-reported memory decline and concern/worry were the sole features associated with high Aβ, but not tau, burden. A higher number of fulfilled SCD-plus features are linked to high Aβ and tau burden. Use of multiple SCD-plus features may help identify early stages of biological AD.

Reliability of structural brain change in cognitively healthy adult samples

In neuroimaging research, tracking individuals over time is key to understanding the interplay between brain changes and genetic, environmental, or cognitive factors across the lifespan. Yet, the extent to which we can estimate the individual trajectories of brain change over time with precision remains uncertain. In this study, we estimated the reliability of structural brain change in cognitively healthy adults from multiple samples and assessed the influence of follow-up time and number of observations. Estimates of cross-sectional measurement error and brain change variance were obtained using the longitudinal FreeSurfer processing stream. Our findings showed, on average, modest longitudinal reliability with two years of follow-up. Increasing the follow-up time was associated with a substantial increase in longitudinal reliability while the impact of increasing the number of observations was comparatively minor. On average, 2-year follow-up studies require ≈2.7 and ≈4.0 times more individuals than designs with follow-ups of 4 and 6 years to achieve comparable statistical power. Subcortical volume exhibited higher longitudinal reliability compared to cortical area, thickness, and volume. The reliability estimates were comparable to those estimated from empirical data. The reliability estimates were affected by both the cohort's age where younger adults had lower reliability of change, and the preprocessing pipeline where the FreeSurfer's longitudinal stream was notably superior than the cross-sectional. Suboptimal reliability inflated sample size requirements and compromised the ability to distinguish individual trajectories of brain aging. This study underscores the importance of long-term follow-ups and the need to consider reliability in longitudinal neuroimaging research.

A review of AI-based radiogenomics in neurodegenerative disease

Neurodegenerative diseases are chronic, progressive conditions that cause irreversible damage to the nervous system, particularly in aging populations. Early diagnosis is a critical challenge, as these diseases often develop slowly and without clear symptoms until significant damage has occurred. Recent advances in radiomics and genomics have provided valuable insights into the mechanisms of these diseases by identifying specific imaging features and genomic patterns. Radiogenomics enhances diagnostic capabilities by linking genomics with imaging phenotypes, offering a more comprehensive understanding of disease progression. The growing field of artificial intelligence (AI), including machine learning and deep learning, opens new opportunities for improving the accuracy and timeliness of these diagnoses. This review examines the application of AI-based radiogenomics in neurodegenerative diseases, summarizing key model designs, performance metrics, publicly available data resources, significant findings, and future research directions. It provides a starting point and guidance for those seeking to explore this emerging area of study.

Contrastive self-supervised learning for neurodegenerative disorder classification

Introduction

Neurodegenerative diseases such as Alzheimer's disease (AD) or frontotemporal lobar degeneration (FTLD) involve specific loss of brain volume, detectable in vivo using T1-weighted MRI scans. Supervised machine learning approaches classifying neurodegenerative diseases require diagnostic-labels for each sample. However, it can be difficult to obtain expert labels for a large amount of data. Self-supervised learning (SSL) offers an alternative for training machine learning models without data-labels.

Methods

We investigated if the SSL models can be applied to distinguish between different neurodegenerative disorders in an interpretable manner. Our method comprises a feature extractor and a downstream classification head. A deep convolutional neural network, trained with a contrastive loss, serves as the feature extractor that learns latent representations. The classification head is a single-layer perceptron that is trained to perform diagnostic group separation. We used N = 2,694 T1-weighted MRI scans from four data cohorts: two ADNI datasets, AIBL and FTLDNI, including cognitively normal controls (CN), cases with prodromal and clinical AD, as well as FTLD cases differentiated into its phenotypes.

Results

Our results showed that the feature extractor trained in a self-supervised way provides generalizable and robust representations for the downstream classification. For AD vs. CN, our model achieves 82% balanced accuracy on the test subset and 80% on an independent holdout dataset. Similarly, the Behavioral variant of frontotemporal dementia (BV) vs. CN model attains an 88% balanced accuracy on the test subset. The average feature attribution heatmaps obtained by the Integrated Gradient method highlighted hallmark regions, i.e., temporal gray matter atrophy for AD, and insular atrophy for BV.

Conclusion

Our models perform comparably to state-of-the-art supervised deep learning approaches. This suggests that the SSL methodology can successfully make use of unannotated neuroimaging datasets as training data while remaining robust and interpretable.

Digital detector PET/CT increases Centiloid measures of amyloid in Alzheimer's disease: A head-to-head comparison of cameras

BACKGROUND

The introduction of therapeutics for Alzheimer's disease has led to increased interest in precisely quantifying amyloid-β (Aβ) burden for diagnosis, treatment monitoring, and further clinical research. Recent positron emission tomography (PET) hardware innovations including digital detectors have led to superior resolution and sensitivity, improving quantitative accuracy. However, the effect of PET scanner on Centiloid remains relatively unexplored and is assumed to be minimized by harmonizing PET resolutions.

OBJECTIVE

To quantify the differences in Centiloid between scanners in a paired cohort.

METHODS

36 participants from the Australian Imaging, Biomarker and Lifestyle study (AIBL) cohort were scanned within a year on two scanners. Each participant underwent 18F-NAV4694 imaging on two of the three scanners investigated, the Siemens Vision, the Siemens mCT and the Philips Gemini. We compared Aβ Centiloid quantification between scanners and assessed the effectiveness of post-reconstruction PET resolution harmonization. We further compared the scanner differences in target sub-regions and with different reference regions to assess spatial variability.

RESULTS

Centiloid from the Vision camera was found to be significantly higher compared to the Gemini and mCT; the difference was greater at high-Centiloid levels. Post-reconstruction resolution harmonization only accounted for and corrected ∼20% of the Centiloid (CL) difference between scanners. We further demonstrated that residual differences have effects that vary spatially between different subregions of the Centiloid mask.

CONCLUSIONS

We have demonstrated that the type of PET scanner that a participant is scanned on affects Centiloid quantification, even when scanner resolution is harmonized. We conclude by highlighting the need for further investigation into harmonization techniques that consider scanner differences.

A Neural Network Approach to Identify Left-Right Orientation of Anatomical Brain MRI

PURPOSE

This study presents a novel application of deep learning to enhance the accuracy of left-right orientation identification in anatomical brain MRI scans. Left-right orientation misidentification in brain MRIs presents significant challenges due to several factors, including metadata loss or ambiguity, which often occurs during the de-identification of medical images for research, conversion between image formats, software operations that strip or overwrite metadata, and the use of older imaging systems that stored orientation differently.

METHOD

A three-dimensional convolutional neural network model was trained using 350 MRIs and evaluated on the basis of eight distinct brain MRI databases, totaling 3056 MRIs, to assess its performance across various conditions, including neurodegenerative diseases.

FINDING

The proposed deep-learning framework demonstrated a 99.8% accuracy in identifying the left-right orientation, thus, addressing challenges associated with the loss of orientation metadata. GradCAM was used to visualize areas of the brain where the model focused, demonstrating the importance of the right planum temporale and surrounding areas in judging left-right orientation. The planum temporale is known to exhibit notable left-right asymmetry related to language functions, underscoring the biological validity of the model. The half of the four left-right misidentified MRIs involved notable brain feature variations, such as a large arachnoidal cyst adjacent to the temporal lobe or ventricular asymmetry, indicating areas for further investigation.

CONCLUSION

This approach offers a potential solution to the persistent issue of left-right misorientation in brain MRIs and supports the reliability of neuroscientific research by ensuring accurate data interpretation.

Grey-Matter Structure Markers of Alzheimer's Disease, Alzheimer's Conversion, Functioning and Cognition: A Meta-Analysis Across 11 Cohorts

Alzheimer's disease (AD) brain markers are needed to select people with early-stage AD for clinical trials and as quantitative endpoint measures in trials. Using 10 clinical cohorts (N = 9140) and the community volunteer UK Biobank (N = 37,664) we performed region of interest (ROI) and vertex-wise analyses of grey-matter structure (thickness, surface area and volume). We identified 94 trait-ROI significant associations, and 307 distinct cluster of vertex-associations, which partly overlap the ROI associations. For AD versus controls, smaller hippocampus, amygdala and of the medial temporal lobe (fusiform and parahippocampal gyri) was confirmed and the vertex-wise results provided unprecedented localisation of some of the associated region. We replicated AD associated differences in several subcortical (putamen, accumbens) and cortical regions (inferior parietal, postcentral, middle temporal, transverse temporal, inferior temporal, paracentral, superior frontal). These grey-matter regions and their relative effect sizes can help refine our understanding of the brain regions that may drive or precede the widespread brain atrophy observed in AD. An AD grey-matter score evaluated in independent cohorts was significantly associated with cognition, MCI status, AD conversion (progression from cognitively normal or MCI to AD), genetic risk, and tau concentration in individuals with none or mild cognitive impairments (AUC in 0.54-0.70, p-value < 5e-4). In addition, some of the grey-matter regions associated with cognitive impairment, progression to AD ('conversion'), and cognition/functional scores were also associated with AD, which sheds light on the grey-matter markers of disease stages, and their relationship with cognitive or functional impairment. Our multi-cohort approach provides robust and fine-grained maps the grey-matter structures associated with AD, symptoms, and progression, and calls for even larger initiatives to unveil the full complexity of grey-matter structure in AD.

Developing multifactorial dementia prediction models using clinical variables from cohorts in the US and Australia

Existing dementia prediction models using non-neuroimaging clinical measures have been limited in their ability to identify disease. This study used machine learning to re-examine the diagnostic potential of clinical measures for dementia. Data was sourced from the Australian Imaging, Biomarkers, and Lifestyle Flagship Study of Ageing (AIBL) and the Alzheimer's Disease Neuroimaging Initiative (ADNI). Clinical variables included 21 measures across medical history, hematological and other blood tests, and APOE genotype. Tree-based machine learning algorithms and artificial neural networks were used. APOE genotype was the best predictor of dementia cases and healthy controls. Our results, however, demonstrated that there are limitations when using publicly accessible cohort data that may limit the generalizability and interpretability of such predictive models. Future research should examine the use of routine APOE genetic testing for dementia diagnostics. It should also focus on clearly unifying data across clinical cohorts.

DeepResBat: Deep residual batch harmonization accounting for covariate distribution differences

Pooling MRI data from multiple datasets requires harmonization to reduce undesired inter-site variabilities, while preserving effects of biological variables (or covariates). The popular harmonization approach ComBat uses a mixed effect regression framework that explicitly accounts for covariate distribution differences across datasets. There is also significant interest in developing harmonization approaches based on deep neural networks (DNNs), such as conditional variational autoencoder (cVAE). However, current DNN approaches do not explicitly account for covariate distribution differences across datasets. Here, we provide mathematical results, suggesting that not accounting for covariates can lead to suboptimal harmonization. We propose two DNN-based covariate-aware harmonization approaches: covariate VAE (coVAE) and DeepResBat. The coVAE approach is a natural extension of cVAE by concatenating covariates and site information with site- and covariate-invariant latent representations. DeepResBat adopts a residual framework inspired by ComBat. DeepResBat first removes the effects of covariates with nonlinear regression trees, followed by eliminating site differences with cVAE. Finally, covariate effects are added back to the harmonized residuals. Using three datasets from three continents with a total of 2787 participants and 10,085 anatomical T1 scans, we find that DeepResBat and coVAE outperformed ComBat, CovBat and cVAE in terms of removing dataset differences, while enhancing biological effects of interest. However, coVAE hallucinates spurious associations between anatomical MRI and covariates even when no association exists. Future studies proposing DNN-based harmonization approaches should be aware of this false positive pitfall. Overall, our results suggest that DeepResBat is an effective deep learning alternative to ComBat. Code for DeepResBat can be found here: https://github.com/ThomasYeoLab/CBIG/tree/master/stable_projects/harmonization/An2024_DeepResBat.

IGUANe: A 3D generalizable CycleGAN for multicenter harmonization of brain MR images

In MRI studies, the aggregation of imaging data from multiple acquisition sites enhances sample size but may introduce site-related variabilities that hinder consistency in subsequent analyses. Deep learning methods for image translation have emerged as a solution for harmonizing MR images across sites. In this study, we introduce IGUANe (Image Generation with Unified Adversarial Networks), an original 3D model that leverages the strengths of domain translation and straightforward application of style transfer methods for multicenter brain MR image harmonization. IGUANe extends CycleGAN by integrating an arbitrary number of domains for training through a many-to-one architecture. The framework based on domain pairs enables the implementation of sampling strategies that prevent confusion between site-related and biological variabilities. During inference, the model can be applied to any image, even from an unknown acquisition site, making it a universal generator for harmonization. Trained on a dataset comprising T1-weighted images from 11 different scanners, IGUANe was evaluated on data from unseen sites. The assessments included the transformation of MR images with traveling subjects, the preservation of pairwise distances between MR images within domains, the evolution of volumetric patterns related to age and Alzheimer's disease (AD), and the performance in age regression and patient classification tasks. Comparisons with other harmonization and normalization methods suggest that IGUANe better preserves individual information in MR images and is more suitable for maintaining and reinforcing variabilities related to age and AD. Future studies may further assess IGUANe in other multicenter contexts, either using the same model or retraining it for applications to different image modalities. Codes and the trained IGUANe model are available at https://github.com/RocaVincent/iguane_harmonization.git.

Clinical validity of IntelliSpace Cognition digital assessment platform in mild cognitive impairment

We evaluated a digital cognitive assessment platform, Philips IntelliSpace Cognition, in a case-control study of patients diagnosed with mild cognitive impairment (MCI) and cognitively normal (CN) older adults. Performance on individual neuropsychological tests, cognitive z-scores, and Alzheimer's disease (AD)-specific composite scores was compared between the CN and MCI groups. These groups were matched for age, sex, and education. Performance on all but two neuropsychological tests was worse in the MCI group. After ranking the cognitive scores by effect size, we found that the memory score was the most impaired, followed by executive functioning. The Early AD/MCI Alzheimer's Cognitive Composite (EMACC) and Preclinical Alzheimer's Cognitive Composite (PACC) scores were constructed from the digital tests on Philips IntelliSpace Cognition. Both AD-specific composite scores showed greater sensitivity and specificity than the Mini-Mental State Examination or individual cognitive z-scores. Together, these results demonstrate the diagnostic value of Philips IntelliSpace Cognition in patients with MCI.

SHIVA-CMB: a deep-learning-based robust cerebral microbleed segmentation tool trained on multi-source T2*GRE- and susceptibility-weighted MRI

Cerebral microbleeds (CMB) represent a feature of cerebral small vessel disease (cSVD), a prominent vascular contributor to age-related cognitive decline, dementia, and stroke. They are visible as spherical hypointense signals on T2*- or susceptibility-weighted magnetic resonance imaging (MRI) sequences. An increasing number of automated CMB detection methods being proposed are based on supervised deep learning (DL). Yet, the lack of open sharing of pre-trained models hampers the practical application and evaluation of these methods beyond specific data sources used in each study. Here, we present the SHIVA-CMB detector, a 3D Unet-based tool trained on 450 scans taken from seven acquisitions in six different cohort studies that included both T2*- and susceptibility-weighted MRI. In a held-out test set of 96 scans, it achieved the sensitivity, precision, and F1 (or Dice similarity coefficient) score of 0.67, 0.82, and 0.74, with less than one false positive detection per image (FPavg = 0.6) and per CMB (FPcmb = 0.15). It achieved a similar level of performance in a separate, evaluation-only dataset with acquisitions never seen during the training (0.67, 0.91, 0.77, 0.5, 0.07 for the sensitivity, precision, F1 score, FPavg, and FPcmb). Further demonstrating its generalizability, it showed a high correlation (Pearson's R = 0.89, p < 0.0001) with a visual count by expert raters in another independent set of 1992 T2*-weighted scans from a large, multi-center cohort study. Importantly, we publicly share both the pipeline ( https://github.com/pboutinaud/SHiVAi/ ) and pre-trained models ( https://github.com/pboutinaud/SHIVA-CMB/ ) to the research community to promote the active application and evaluation of our tool. We believe this effort will help accelerate research on the pathophysiology and functional consequences of CMB by enabling rapid characterization of CMB in large-scale studies.

Brain change trajectories in healthy adults correlate with Alzheimer's related genetic variation and memory decline across life

Throughout adulthood and ageing our brains undergo structural loss in an average pattern resembling faster atrophy in Alzheimer's disease (AD). Using a longitudinal adult lifespan sample (aged 30-89; 2-7 timepoints) and four polygenic scores for AD, we show that change in AD-sensitive brain features correlates with genetic AD-risk and memory decline in healthy adults. We first show genetic risk links with more brain loss than expected for age in early Braak regions, and find this extends beyond APOE genotype. Next, we run machine learning on AD-control data from the Alzheimer's Disease Neuroimaging Initiative using brain change trajectories conditioned on age, to identify AD-sensitive features and model their change in healthy adults. Genetic AD-risk linked with multivariate change across many AD-sensitive features, and we show most individuals over age ~50 are on an accelerated trajectory of brain loss in AD-sensitive regions. Finally, high genetic risk adults with elevated brain change showed more memory decline through adulthood, compared to high genetic risk adults with less brain change. Our findings suggest quantitative AD risk factors are detectable in healthy individuals, via a shared pattern of ageing- and AD-related neurodegeneration that occurs along a continuum and tracks memory decline through adulthood.

Multimodal Alzheimer's disease classification through ensemble deep random vector functional link neural network

Alzheimer's disease (AD) is a condition with a complex pathogenesis, sometimes hereditary, characterized by the loss of neurons and synapses, along with the presence of senile plaques and neurofibrillary tangles. Early detection, particularly among individuals at high risk, is critical for effective treatment or prevention, yet remains challenging due to data variability and incompleteness. Most current research relies on single data modalities, potentially limiting comprehensive staging of AD. This study addresses this gap by integrating multimodal data-including clinical and genetic information-using deep learning (DL) models, with a specific focus on random vector functional link (RVFL) networks, to enhance early detection of AD and mild cognitive impairment (MCI). Our findings demonstrate that ensemble deep RVFL (edRVFL) models, when combined with effective data imputation techniques such as Winsorized-mean (Wmean), achieve superior performance in detecting early stages of AD. Notably, the edRVFL model achieved an accuracy of 98.8%, precision of 98.3%, recall of 98.4%, and F1-score of 98.2%, outperforming traditional machine learning models like support vector machines, random forests, and decision trees. This underscores the importance of integrating advanced imputation strategies and deep learning techniques in AD diagnosis.

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.

A systematic review of generalization research in medical image classification

Numerous Deep Learning (DL) classification models have been developed for a large spectrum of medical image analysis applications, which promises to reshape various facets of medical practice. Despite early advances in DL model validation and implementation, which encourage healthcare institutions to adopt them, a fundamental questions remain: how can these models effectively handle domain shift? This question is crucial to limit DL models performance degradation. Medical data are dynamic and prone to domain shift, due to multiple factors. Two main shift types can occur over time: (1) covariate shift mainly arising due to updates to medical equipment and (2) concept shift caused by inter-grader variability. To mitigate the problem of domain shift, existing surveys mainly focus on domain adaptation techniques, with an emphasis on covariate shift. More generally, no work has reviewed the state-of-the-art solutions while focusing on the shift types. This paper aims to explore existing domain generalization methods for DL-based classification models through a systematic review of literature. It proposes a taxonomy based on the shift type they aim to solve. Papers were searched and gathered on Scopus till 10 April 2023, and after the eligibility screening and quality evaluation, 77 articles were identified. Exclusion criteria included: lack of methodological novelty (e.g., reviews, benchmarks), experiments conducted on a single mono-center dataset, or articles not written in English. The results of this paper show that learning based methods are emerging, for both shift types. Finally, we discuss future challenges, including the need for improved evaluation protocols and benchmarks, and envisioned future developments to achieve robust, generalized models for medical image classification.

Evaluation of Brain Age as a Specific Marker of Brain Health

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

The interplay of age, gender and amyloid on brain and cognition in mid-life and older adults

Deficits in memory are seen as a canonical sign of aging and a prodrome to dementia in older adults. However, our understanding of age-related cognition and brain morphology occurring throughout a broader spectrum of adulthood remains limited. We quantified the relationship between cognitive function and brain morphology (sulcal width, SW) using three cross-sectional observational datasets (PISA, AIBL, ADNI) from mid-life to older adulthood, assessing the influence of age, sex, amyloid (Aβ) and genetic risk for dementia. The data comprised cognitive, genetic and neuroimaging measures of a total of 1570 non-clinical mid-life and older adults (mean age 72, range 49-90 years, 1330 males) and 1365 age- and sex-matched adults with mild cognitive impairment (MCI) or Alzheimer's disease (AD). Among non-clinical adults, we found robust modes of co-variation between regional SW and multidomain cognitive function that differed between the mid-life and older age range. These cortical and cognitive profiles derived from healthy cohorts predicted out-of-sample AD and MCI. Furthermore, Aβ-deposition and educational attainment levels were associated with cognition but not SW. These findings underscoring the complex interplay between factors influencing cognition and brain structure from mid-life onwards, providing valuable insights for future research into neurodegeneration and the development of future screening algorithms.

AI for the prediction of early stages of Alzheimer's disease from neuroimaging biomarkers - A narrative review of a growing field

OBJECTIVES

The objectives of this narrative review are to summarize the current state of AI applications in neuroimaging for early Alzheimer's disease (AD) prediction and to highlight the potential of AI techniques in improving early AD diagnosis, prognosis, and management.

METHODS

We conducted a narrative review of studies using AI techniques applied to neuroimaging data for early AD prediction. We examined single-modality studies using structural MRI and PET imaging, as well as multi-modality studies integrating multiple neuroimaging techniques and biomarkers. Furthermore, they reviewed longitudinal studies that model AD progression and identify individuals at risk of rapid decline.

RESULTS

Single-modality studies using structural MRI and PET imaging have demonstrated high accuracy in classifying AD and predicting progression from mild cognitive impairment (MCI) to AD. Multi-modality studies, integrating multiple neuroimaging techniques and biomarkers, have shown improved performance and robustness compared to single-modality approaches. Longitudinal studies have highlighted the value of AI in modeling AD progression and identifying individuals at risk of rapid decline. However, challenges remain in data standardization, model interpretability, generalizability, clinical integration, and ethical considerations.

CONCLUSION

AI techniques applied to neuroimaging data have the potential to improve early AD diagnosis, prognosis, and management. Addressing challenges related to data standardization, model interpretability, generalizability, clinical integration, and ethical considerations is crucial for realizing the full potential of AI in AD research and clinical practice. Collaborative efforts among researchers, clinicians, and regulatory agencies are needed to develop reliable, robust, and ethical AI tools that can benefit AD patients and society.

OpenMAP-T1: A Rapid Deep-Learning Approach to Parcellate 280 Anatomical Regions to Cover the Whole Brain

This study introduces OpenMAP-T1, a deep-learning-based method for rapid and accurate whole-brain parcellation in T1- weighted brain MRI, which aims to overcome the limitations of conventional normalization-to-atlas-based approaches and multi-atlas label-fusion (MALF) techniques. Brain image parcellation is a fundamental process in neuroscientific and clinical research, enabling a detailed analysis of specific cerebral regions. Normalization-to-atlas-based methods have been employed for this task, but they face limitations due to variations in brain morphology, especially in pathological conditions. The MALF techniques improved the accuracy of the image parcellation and robustness to variations in brain morphology, but at the cost of high computational demand that requires a lengthy processing time. OpenMAP-T1 integrates several convolutional neural network models across six phases: preprocessing; cropping; skull-stripping; parcellation; hemisphere segmentation; and final merging. This process involves standardizing MRI images, isolating the brain tissue, and parcellating it into 280 anatomical structures that cover the whole brain, including detailed gray and white matter structures, while simplifying the parcellation processes and incorporating robust training to handle various scan types and conditions. The OpenMAP-T1 was validated on the Johns Hopkins University atlas library and eight available open resources, including real-world clinical images, and the demonstration of robustness across different datasets with variations in scanner types, magnetic field strengths, and image processing techniques, such as defacing. Compared with existing methods, OpenMAP-T1 significantly reduced the processing time per image from several hours to less than 90 s without compromising accuracy. It was particularly effective in handling images with intensity inhomogeneity and varying head positions, conditions commonly seen in clinical settings. The adaptability of OpenMAP-T1 to a wide range of MRI datasets and its robustness to various scan conditions highlight its potential as a versatile tool in neuroimaging.

Autoantibodies to BACE1 promote Aβ accumulation and neurodegeneration in Alzheimer's disease

The profile of autoantibodies is dysregulated in patients with Alzheimer's disease (AD). Autoantibodies to beta-site amyloid precursor protein (APP)-cleaving enzyme 1 (BACE1) are present in human blood. This study aims to investigate the clinical relevance and pathophysiological roles of autoantibodies to BACE1 in AD. Clinical investigations were conducted in two independent cohorts, the Chongqing cohort, and the Australian Imaging, Biomarkers, and Lifestyle (AIBL) cohort. The Chongqing cohort included 55 AD patients, 28 patients with non-AD dementia, and 70 cognitively normal subjects (CN). The AIBL cohort included 162 Aβ-PET- CN, 169 Aβ-PET+ cognitively normal subjects (preclinical AD), and 31 Aβ-PET+ cognitively impaired subjects (Clinical AD). Plasma autoantibodies to BACE1 were determined by one-site Elisa. The associations of plasma autoantibodies to BACE1 with brain Aβ load and cognitive trajectory were investigated. The effects of autoantibodies to BACE1 on AD-type pathologies and underlying mechanisms were investigated in APP/PS1 mice and SH/APPswe/PS1wt cell lines. In the Chongqing cohort, plasma autoantibodies to BACE1 were higher in AD patients, in comparison with CN and non-AD dementia patients. In the AIBL cohort, plasma autoantibodies to BACE1 were highest in clinical AD patients, followed by preclinical AD and CN subjects. Higher autoantibodies to BACE1 were associated with an increased incidence of brain amyloid positivity conversion during follow-up. Autoantibodies to BACE1 exacerbated brain amyloid deposition and subsequent AD-type pathologies, including Tau hyperphosphorylation, neuroinflammation, and neurodegeneration in APP/PS1 mice. Autoantibodies to BACE1 increased Aβ production by promoting BACE1 expression through inhibiting PPARγ signaling. These findings suggest that autoantibodies to BACE1 are pathogenic in AD and the upregulation of these autoantibodies may promote the development of the disease. This study offers new insights into the mechanism of AD from an autoimmune perspective.

Deep learning-based approaches for multi-omics data integration and analysis

BACKGROUND

The rapid growth of deep learning, as well as the vast and ever-growing amount of available data, have provided ample opportunity for advances in fusion and analysis of complex and heterogeneous data types. Different data modalities provide complementary information that can be leveraged to gain a more complete understanding of each subject. In the biomedical domain, multi-omics data includes molecular (genomics, transcriptomics, proteomics, epigenomics, metabolomics, etc.) and imaging (radiomics, pathomics) modalities which, when combined, have the potential to improve performance on prediction, classification, clustering and other tasks. Deep learning encompasses a wide variety of methods, each of which have certain strengths and weaknesses for multi-omics integration.

METHOD

In this review, we categorize recent deep learning-based approaches by their basic architectures and discuss their unique capabilities in relation to one another. We also discuss some emerging themes advancing the field of multi-omics integration.

RESULTS

Deep learning-based multi-omics integration methods were categorized broadly into non-generative (feedforward neural networks, graph convolutional neural networks, and autoencoders) and generative (variational methods, generative adversarial models, and a generative pretrained model). Generative methods have the advantage of being able to impose constraints on the shared representations to enforce certain properties or incorporate prior knowledge. They can also be used to generate or impute missing modalities. Recent advances achieved by these methods include the ability to handle incomplete data as well as going beyond the traditional molecular omics data types to integrate other modalities such as imaging data.

CONCLUSION

We expect to see further growth in methods that can handle missingness, as this is a common challenge in working with complex and heterogeneous data. Additionally, methods that integrate more data types are expected to improve performance on downstream tasks by capturing a comprehensive view of each sample.

Brain aging patterns in a large and diverse cohort of 49,482 individuals

Brain aging process is influenced by various lifestyle, environmental and genetic factors, as well as by age-related and often coexisting pathologies. Magnetic resonance imaging and artificial intelligence methods have been instrumental in understanding neuroanatomical changes that occur during aging. Large, diverse population studies enable identifying comprehensive and representative brain change patterns resulting from distinct but overlapping pathological and biological factors, revealing intersections and heterogeneity in affected brain regions and clinical phenotypes. Herein, we leverage a state-of-the-art deep-representation learning method, Surreal-GAN, and present methodological advances and extensive experimental results elucidating brain aging heterogeneity in a cohort of 49,482 individuals from 11 studies. Five dominant patterns of brain atrophy were identified and quantified for each individual by respective measures, R-indices. Their associations with biomedical, lifestyle and genetic factors provide insights into the etiology of observed variances, suggesting their potential as brain endophenotypes for genetic and lifestyle risks. Furthermore, baseline R-indices predict disease progression and mortality, capturing early changes as supplementary prognostic markers. These R-indices establish a dimensional approach to measuring aging trajectories and related brain changes. They hold promise for precise diagnostics, especially at preclinical stages, facilitating personalized patient management and targeted clinical trial recruitment based on specific brain endophenotypic expression and prognosis.

HiMAL: Multimodal Hierarchical Multi-task Auxiliary Learning framework for predicting Alzheimer's disease progression

Objective

We aimed to develop and validate a novel multimodal framework Hierarchical Multi-task Auxiliary Learning (HiMAL) framework, for predicting cognitive composite functions as auxiliary tasks that estimate the longitudinal risk of transition from Mild Cognitive Impairment (MCI) to Alzheimer's Disease (AD).

Materials and Methods

HiMAL utilized multimodal longitudinal visit data including imaging features, cognitive assessment scores, and clinical variables from MCI patients in the Alzheimer's Disease Neuroimaging Initiative dataset, to predict at each visit if an MCI patient will progress to AD within the next 6 months. Performance of HiMAL was compared with state-of-the-art single-task and multitask baselines using area under the receiver operator curve (AUROC) and precision recall curve (AUPRC) metrics. An ablation study was performed to assess the impact of each input modality on model performance. Additionally, longitudinal explanations regarding risk of disease progression were provided to interpret the predicted cognitive decline.

Results

Out of 634 MCI patients (mean [IQR] age: 72.8 [67-78], 60% male), 209 (32%) progressed to AD. HiMAL showed better prediction performance compared to all state-of-the-art longitudinal single-modality singe-task baselines (AUROC = 0.923 [0.915-0.937]; AUPRC = 0.623 [0.605-0.644]; all P < .05). Ablation analysis highlighted that imaging and cognition scores with maximum contribution towards prediction of disease progression.

Discussion

Clinically informative model explanations anticipate cognitive decline 6 months in advance, aiding clinicians in future disease progression assessment. HiMAL relies on routinely collected electronic health records (EHR) variables for proximal (6 months) prediction of AD onset, indicating its translational potential for point-of-care monitoring and managing of high-risk patients.

Prevalence of neurocognitive disorders 5 years after elective orthopaedic surgery

BACKGROUND

Peri-operative neurocognitive disorders are one of the most common complications affecting older adults after anaesthesia and surgery. It is not clear how exposure to surgery and anaesthesia contributes to the prevalence of long-term neurocognitive disorders. This study aimed to report the prevalence of neurocognitive disorders, and explore pre-operative factors associated with neurocognitive disorders 5 years after elective orthopaedic surgery.

METHODS

A prospective, 5-year longitudinal, cohort study was performed recruiting patients (aged ≥ 60 y) undergoing elective orthopaedic surgery and a contemporaneous non-surgical control group. Neurocognitive disorder was evaluated and classified at baseline and 5-year review incorporating: self- and informant-reported cognition; functional participation; and performance on neuropsychological tests.

RESULTS

Recruitment at 5-year follow-up included 195 patients and 21 control participants. In the patient cohort the prevalence of neurocognitive disorder was 38.1% (n = 75), with 61 (30.1%) meeting the criteria for mild neurocognitive disorder and 14 (7.1%) for major neurocognitive disorder. At 5-year follow-up, 121 (61.4%) patients were classified with a neurocognitive disorder, with 88 (44.7%) characterised with mild neurocognitive disorder and 33 (16.8%) with major neurocognitive disorder. Age (odds ratio (95%CI) 1.07 (1.02-1.13); p = 0.01) and baseline cognitive impairment (odds ratio (95%CI) 2.1 (1.06-4.15); p = 0.03) were significant predictors of neurocognitive disorder 5 years after surgery.

CONCLUSION

More than half of older adult patients had some form of neurocognitive disorder 5 years after elective orthopaedic surgery. Surgery and anaesthesia may be associated with the trajectory of cognitive decline in at-risk older adults, including those with pre-operative cognitive impairment. Cognitive screening should be factored into pre-operative assessments of older adults to inform subsequent care.

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

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

Prediction of future dementia among patients with mild cognitive impairment (MCI) by integrating multimodal clinical data

Efficiently and objectively analyzing the complex, diverse multimodal data collected from patients at risk for dementia can be difficult in the clinical setting, contributing to high rates of underdiagnosis or misdiagnosis of this serious disorder. Patients with mild cognitive impairment (MCI) are especially at risk of developing dementia in the future. This study evaluated the ability of multi-modal machine learning (ML) methods, especially the Ensemble Integration (EI) framework, to predict future dementia development among patients with MCI. EI is a machine learning framework designed to leverage complementarity and consensus in multimodal data, which may not be adequately captured by methods used by prior dementia-related prediction studies. We tested EI's ability to predict future dementia development among MCI patients using multimodal clinical and imaging data, such as neuroanatomical measurements from structural magnetic resonance imaging (MRI) and positron emission tomography (PET) scans, from The Alzheimer's Disease Prediction of Longitudinal Evolution (TADPOLE) challenge. For predicting future dementia development among MCI patients, on a held out test set, the EI-based model performed better (AUC = 0.81, F-measure = 0.68) than the more commonly used XGBoost (AUC = 0.68, F-measure = 0.57) and deep learning (AUC = 0.79, F-measure = 0.61) approaches. This EI-based model also suggested MRI-derived volumes of regions in the middle temporal gyrus, posterior cingulate gyrus and inferior lateral ventricle brain regions to be predictive of progression to dementia.

MethylGenotyper: Accurate Estimation of SNP Genotypes and Genetic Relatedness from DNA Methylation Data

Epigenome-wide association studies (EWAS) are susceptible to widespread confounding caused by population structure and genetic relatedness. Nevertheless, kinship estimation is challenging in EWAS without genotyping data. Here, we proposed MethylGenotyper, a method that for the first time enables accurate genotyping at thousands of single nucleotide polymorphisms (SNPs) directly from commercial DNA methylation microarrays. We modeled the intensities of methylation probes near SNPs with a mixture of three beta distributions corresponding to different genotypes and estimated parameters with an expectation-maximization algorithm. We conducted extensive simulations to demonstrate the performance of the method. When applying MethylGenotyper to the Infinium EPIC array data of 4662 Chinese samples, we obtained genotypes at 4319 SNPs with a concordance rate of 98.26%, enabling the identification of 255 pairs of close relatedness. Furthermore, we showed that MethylGenotyper allows for the estimation of both population structure and cryptic relatedness among 702 Australians of diverse ancestry. We also implemented MethylGenotyper in a publicly available R package (https://github.com/Yi-Jiang/MethylGenotyper) to facilitate future large-scale EWAS.

Clearance and transport of amyloid β by peripheral monocytes correlate with Alzheimer's disease progression

Impaired clearance of amyloid β (Aβ) in late-onset Alzheimer's disease (AD) affects disease progression. The role of peripheral monocytes in Aβ clearance from the central nervous system (CNS) is unclear. We use a flow cytometry assay to identify Aβ-binding monocytes in blood, validated by confocal microscopy, Western blotting, and mass spectrometry. Flow cytometry immunophenotyping and correlation with AD biomarkers are studied in 150 participants from the AIBL study. We also examine monocytes in human cerebrospinal fluid (CSF) and their migration in an APP/PS1 mouse model. The assay reveals macrophage-like Aβ-binding monocytes with high phagocytic potential in both the periphery and CNS. We find lower surface Aβ levels in mild cognitive impairment (MCI) and AD-dementia patients compared to cognitively unimpaired individuals. Monocyte infiltration from blood to CSF and migration from CNS to peripheral lymph nodes and blood are observed. Here we show that Aβ-binding monocytes may play a role in CNS Aβ clearance, suggesting their potential as a biomarker for AD diagnosis and monitoring.

3T sodium-MRI as predictor of neurocognition in nondemented older adults: a cross sectional study

Dementia is a burgeoning global problem. Novel magnetic resonance imaging (MRI) metrics beyond volumetry may bring new insight and aid clinical trial evaluation of interventions early in the Alzheimer's disease course to complement existing imaging and clinical metrics. To determine whether: (i) normalized regional sodium-MRI values (Na-SI) are better predictors of neurocognitive status than volumetry (ii) cerebral amyloid PET status improves modelling. Nondemented older adult (>60 years) volunteers of known Alzheimer's Disease Assessment Scale (ADAS-Cog11), Mini-Mental State Examination (MMSE) and Consortium to Establish a Registry for Alzheimer's Disease (CERAD) neurocognitive test scores, ApolipoproteinE (APOE) e4 +/- cerebral amyloid PET status were prospectively recruited for 3T sodium-MRI brain scans. Left and right hippocampal, entorhinal and precuneus volumes and Na-SI (using the proportional intensity scaling normalization method with field inhomogeneity and partial volume corrections) were obtained after segmentation and co-registration of 3D-T1-weighted proton images. Descriptive statistics, correlation and best-subset regression analyses were performed. In our 76 nondemented participants (mean(standard deviation) age 75(5) years; woman 47(62%); cognitively unimpaired 54/76(71%), mildly cognitively impaired 22/76(29%)), left hippocampal Na-SI, not volume, was preferentially in the best models for predicting MMSE (Odds Ratio (OR) = 0.19(Confidence Interval (CI) = 0.07,0.53), P-value = 0.001) and ADAS-Cog11 (Beta(B) = 1.2(CI = 0.28,2.1), P-value = 0.01) scores. In the entorhinal analysis, right entorhinal Na-SI, not volume, was preferentially selected in the best model for predicting ADAS-Cog11 (B = 0.94(CI = 0.11,1.8), P-value = 0.03). While right entorhinal Na-SI and volume were both selected for MMSE modelling (Na-SI OR = 0.23(CI = 0.09,0.6), P-value = 0.003; volume OR = 2.6(CI = 1.0,6.6), P-value = 0.04), independently, Na-SI explained more of the variance (Na-SI R 2 = 10.3; volume R 2 = 7.5). No imaging variable was selected in the best CERAD models. Adding cerebral amyloid status improved model fit (Akaike Information Criterion increased 2.0 for all models, P-value < 0.001-0.045). Regional Na-SI were more predictive of MMSE and ADAS-Cog11 scores in our nondemented older adult cohort than volume, hippocampal more robust than entorhinal region of interest. Positive amyloid status slightly further improved model fit.

Comorbidity-based framework for Alzheimer's disease classification using graph neural networks

Alzheimer's disease (AD), the most prevalent form of dementia, requires early prediction for timely intervention. Current deep learning approaches, particularly those using traditional neural networks, face challenges such as handling high-dimensional data, interpreting complex relationships, and managing data bias. To address these limitations, we propose a framework utilizing graph neural networks (GNNs), which excel in modeling relationships within graph-structured data. Our study employs GNNs on data from the Alzheimer's Disease Neuroimaging Initiative for binary and multi-class classification across the three stages of AD: cognitively normal (CN), mild cognitive impairment (MCI), and Alzheimer's disease (AD). By incorporating comorbidity data derived from electronic health records, we achieved the most effective multi-classification results. Notably, the GNN model (Chebyshev Convolutional Neural Networks) demonstrated superior performance with a 0.98 accuracy in multi-class classification and 0.99, 0.93, and 0.94 in the AD/CN, AD/MCI, and CN/MCI binary tasks, respectively. The model's robustness was further validated using the Australian Imaging, Biomarker & Lifestyle dataset as an external validation set. This work contributes to the field by offering a robust, accurate, and cost-effective method for early AD prediction (CN vs. MCI), addressing key challenges in existing deep learning approaches.

Elucidating the Link Between Anxiety/Depression and Alzheimer's Dementia in the Australian Imaging Biomarkers and Lifestyle (AIBL) Study

BACKGROUND

The associations between mood disorders (anxiety and depression) and mild cognitive impairment (MCI) or Alzheimer's dementia (AD) remain unclear.

METHODS

Data from the Australian Imaging, Biomarker & Lifestyle (AIBL) study were subjected to logistic regression to determine both cross-sectional and longitudinal associations between anxiety/depression and MCI/AD. Effect modification by selected covariates was analysed using the likelihood ratio test.

RESULTS

Cross-sectional analysis was performed to explore the association between anxiety/depression and MCI/AD among 2,209 participants with a mean [SD] age of 72.3 [7.4] years, of whom 55.4% were female. After adjusting for confounding variables, we found a significant increase in the odds of AD among participants with two mood disorders (anxiety: OR 1.65 [95% CI 1.04-2.60]; depression: OR 1.73 [1.12-2.69]). Longitudinal analysis was conducted to explore the target associations among 1,379 participants with a mean age of 71.2 [6.6] years, of whom 56.3% were female. During a mean follow-up of 5.0 [4.2] years, 163 participants who developed MCI/AD (refer to as PRO) were identified. Only anxiety was associated with higher odds of PRO after adjusting for covariates (OR 1.56 [1.03-2.39]). However, after additional adjustment for depression, the association became insignificant. Additionally, age, sex, and marital status were identified as effect modifiers for the target associations.

CONCLUSION

Our study provides supportive evidence that anxiety and depression impact on the evolution of MCI/AD, which provides valuable epidemiological insights that can inform clinical practice, guiding clinicians in offering targeted dementia prevention and surveillance programs to the at-risk populations.

3D Multimodal Fusion Network With Disease-Induced Joint Learning for Early Alzheimer's Disease Diagnosis

Multimodal neuroimaging provides complementary information critical for accurate early diagnosis of Alzheimer's disease (AD). However, the inherent variability between multimodal neuroimages hinders the effective fusion of multimodal features. Moreover, achieving reliable and interpretable diagnoses in the field of multimodal fusion remains challenging. To address them, we propose a novel multimodal diagnosis network based on multi-fusion and disease-induced learning (MDL-Net) to enhance early AD diagnosis by efficiently fusing multimodal data. Specifically, MDL-Net proposes a multi-fusion joint learning (MJL) module, which effectively fuses multimodal features and enhances the feature representation from global, local, and latent learning perspectives. MJL consists of three modules, global-aware learning (GAL), local-aware learning (LAL), and outer latent-space learning (LSL) modules. GAL via a self-adaptive Transformer (SAT) learns the global relationships among the modalities. LAL constructs local-aware convolution to learn the local associations. LSL module introduces latent information through outer product operation to further enhance feature representation. MDL-Net integrates the disease-induced region-aware learning (DRL) module via gradient weight to enhance interpretability, which iteratively learns weight matrices to identify AD-related brain regions. We conduct the extensive experiments on public datasets and the results confirm the superiority of our proposed method. Our code will be available at: https://github.com/qzf0320/MDL-Net.

Lower Prevalence of Asymptomatic Alzheimer's Disease Among Healthy African Americans

OBJECTIVE

Alzheimer's disease (AD) is believed to be more common in African Americans (AA), but biomarker studies in AA populations are limited. This report represents the largest study to date examining cerebrospinal fluid AD biomarkers in AA individuals.

METHODS

We analyzed 3,006 cerebrospinal fluid samples from controls, AD cases, and non-AD cases, including 495 (16.5%) self-identified black/AA and 2,456 (81.7%) white/European individuals using cutoffs derived from the Alzheimer's Disease Neuroimaging Initiative, and using a data-driven multivariate Gaussian mixture of regressions.

RESULTS

Distinct effects of race were found in different groups. Total Tauand phospho181-Tau were lower among AA individuals in all groups (p < 0.0001), and Aβ42 was markedly lower in AA controls compared with white controls (p < 0.0001). Gaussian mixture of regressions modeling of cerebrospinal fluid distributions incorporating adjustments for covariates revealed coefficient estimates for AA race comparable with 2-decade change in age. Using Alzheimer's Disease Neuroimaging Initiative cutoffs, fewer AA controls were classified as biomarker-positive asymptomatic AD (8.0% vs 13.4%). After adjusting for covariates, our Gaussian mixture of regressions model reduced this difference, but continued to predict lower prevalence of asymptomatic AD among AA controls (9.3% vs 13.5%).

INTERPRETATION

Although the risk of dementia is higher, data-driven modeling indicates lower frequency of asymptomatic AD in AA controls, suggesting that dementia among AA populations may not be driven by higher rates of AD. ANN NEUROL 2024;96:463-475.

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.

A reproducible approach for the use of aptamer libraries for the identification of Aptamarkers for brain amyloid deposition based on plasma analysis

An approach for the agnostic identification and validation of aptamers for the prediction of a medical state from plasma analysis is presented in application to a key risk factor for Alzheimer's disease. brain amyloid deposition. This method involved the use of a newly designed aptamer library with sixteen random nucleotides interspersed with fixed sequences called a Neomer library. The Neomer library approach enables the direct application of the same starting library on multiple plasma samples, without the requirement for pre-enrichment associated with the traditional approach. Eight aptamers were identified as a result of the selection process and screened across 390 plasma samples by qPCR assay. Results were analysed using multiple machine learning algorithms from the Scikit-learn package along with clinical variables including cognitive status, age and sex to create predictive models. An Extra Trees Classifier model provided the highest predictive power. The Neomer approach resulted in a sensitivity of 0.88. specificity of 0.76. and AUC of 0.79. The only clinical variables that were included in the model were age and sex. We conclude that the Neomer approach represents a clear improvement for the agnostic identification of aptamers (Aptamarkers) that bind to unknown biomarkers of a medical state.

Alzheimer's disease genetic risk and changes in brain atrophy and white matter hyperintensities in cognitively unimpaired adults

Reduced brain volumes and more prominent white matter hyperintensities on MRI scans are commonly observed among older adults without cognitive impairment. However, it remains unclear whether rates of change in these measures among cognitively normal adults differ as a function of genetic risk for late-onset Alzheimer's disease, including APOE-ɛ4, APOE-ɛ2 and Alzheimer's disease polygenic risk scores (AD-PRS), and whether these relationships are influenced by other variables. This longitudinal study examined the trajectories of regional brain volumes and white matter hyperintensities in relationship to APOE genotypes (N = 1541) and AD-PRS (N = 1093) in a harmonized dataset of middle-aged and older individuals with normal cognition at baseline (mean baseline age = 66 years, SD = 9.6) and an average of 5.3 years of MRI follow-up (max = 24 years). Atrophy on volumetric MRI scans was quantified in three ways: (i) a composite score of regions vulnerable to Alzheimer's disease (SPARE-AD); (ii) hippocampal volume; and (iii) a composite score of regions indexing advanced non-Alzheimer's disease-related brain aging (SPARE-BA). Global white matter hyperintensity volumes were derived from fluid attenuated inversion recovery (FLAIR) MRI. Using linear mixed effects models, there was an APOE-ɛ4 gene-dose effect on atrophy in the SPARE-AD composite and hippocampus, with greatest atrophy among ɛ4/ɛ4 carriers, followed by ɛ4 heterozygouts, and lowest among ɛ3 homozygouts and ɛ2/ɛ2 and ɛ2/ɛ3 carriers, who did not differ from one another. The negative associations of APOE-ɛ4 with atrophy were reduced among those with higher education (P < 0.04) and younger baseline ages (P < 0.03). Higher AD-PRS were also associated with greater atrophy in SPARE-AD (P = 0.035) and the hippocampus (P = 0.014), independent of APOE-ɛ4 status. APOE-ɛ2 status (ɛ2/ɛ2 and ɛ2/ɛ3 combined) was not related to baseline levels or atrophy in SPARE-AD, SPARE-BA or the hippocampus, but was related to greater increases in white matter hyperintensities (P = 0.014). Additionally, there was an APOE-ɛ4 × AD-PRS interaction in relation to white matter hyperintensities (P = 0.038), with greater increases in white matter hyperintensities among APOE-ɛ4 carriers with higher AD-PRS. APOE and AD-PRS associations with MRI measures did not differ by sex. These results suggest that APOE-ɛ4 and AD-PRS independently and additively influence longitudinal declines in brain volumes sensitive to Alzheimer's disease and synergistically increase white matter hyperintensity accumulation among cognitively normal individuals. Conversely, APOE-ɛ2 primarily influences white matter hyperintensity accumulation, not brain atrophy. Results are consistent with the view that genetic factors for Alzheimer's disease influence atrophy in a regionally specific manner, likely reflecting preclinical neurodegeneration, and that Alzheimer's disease risk genes contribute to white matter hyperintensity formation.

Early Alzheimer's Disease Detection: A Review of Machine Learning Techniques for Forecasting Transition from Mild Cognitive Impairment

Alzheimer's disease is a weakening neurodegenerative condition with profound cognitive implications, making early and accurate detection crucial for effective treatment. In recent years, machine learning, particularly deep learning, has shown significant promise in detecting mild cognitive impairment to Alzheimer's disease conversion. This review synthesizes research on machine learning approaches for predicting conversion from mild cognitive impairment to Alzheimer's disease dementia using magnetic resonance imaging, positron emission tomography, and other biomarkers. Various techniques used in literature such as machine learning, deep learning, and transfer learning were examined in this study. Additionally, data modalities and feature extraction methods analyzed by different researchers are discussed. This review provides a comprehensive overview of the current state of research in Alzheimer's disease detection and highlights future research directions.