Cognitively Normal Individuals Research Articles

INTRODUCTIONIt remains unclear whether diabetes mellitus (DM) is associated with Alzheimer's disease (AD) pathology and associated vascular burden.METHODSWe included cognitively normal (CN), mild cognitive impairment (MCI), and dementia individuals. We assessed associations between DM and AD biomarkers (amyloid beta [Aβ], phosphorylated tau‐181 [p‐tau181], total tau [t‐tau], and medial temporal atrophy [MTA]) and vascular burden (white matter hyperintensities, microbleeds) by logistic regression. Secondary analyses assessed associations between DM and profiles of Aβ combined with p‐tau181/t‐tau/MTA/white matter hyperintensity/microbleeds.RESULTSWe included 5550 participants (65.8+‐8.7 years, 8.7% DM). DM was associated with lower odds of abnormal AD biomarkers: Aβ in MCI (odds ratio [OR] = 0.70, 95% confidence interval [CI]: 0.51–0.95, p = 0.02) and dementia (OR = 0.44, 0.26–0.78, p = 0.003), and p‐tau181 in dementia (OR = 0.64, 0.41–1.00, p = 0.045). Secondary analyses indicated associations of DM with abnormal t‐tau (OR = 1.57, 1.00–2.46, p = 0.048) and MTA (OR = 1.96, 1.05–3.68, p = 0.04) only in CN individuals with normal Aβ.DISCUSSIONIn cognitively impaired individuals, DM was associated with lower odds of Aβ pathology, whereas DM was associated with neurodegeneration markers in CN individuals without Aβ pathology.HighlightsDiabetes mellitus (DM) was associated with lower odds of amyloid beta (Aβ) and phosphorylated tau (p‐tau) pathology across clinical populations.DM was associated with total tau and medial temporal atrophy in cognitively normal individuals without Aβ pathology.DM may be associated with dementia through neurodegenerative pathways other than Alzheimer's disease.

Identifying robust neuroimaging biomarkers for Alzheimer’s disease (AD) and mild cognitive impairment (MCI) is essential for early diagnosis and intervention. In this study, we introduce a novel, fully automated, guided dynamic functional connectivity (dFNC) framework for extracting multiple dynamic measures to distinguish MCI/AD from cognitively normal (CN) individuals.Resting-state fMRI data were used to extract subject-specific brain networks via spatially constrained independent component analysis (scICA), using a multi-objective optimization framework to ensure alignment with known functional networks while preserving individual variability. Using these components, dFNC was computed through a sliding-window approach. ICA was then applied to the concatenated dFNC matrices from the UK Biobank (UKBB) dataset to identify five canonical brain states, each representing a replicable, independent pattern of connectivity. These states served as biologically informed priors in a state-constrained ICA (St-cICA), which was applied to each subject in the combined OASIS-3 and ADNI datasets to guide individual-level decomposition and ensure interpretable connectivity states guided by state priors derived from the normative UKBB sample.St-cICA extracted subject-specific dFNC features and associated weighted timecourses. To characterize dFNC patterns, we computed metrics from the most strongly expressed (primary) state and introduced estimation of the second-most expressed (secondary) state at each time point, including dwell time, occupancy rate, and transition probabilities. Group comparisons using two-sample t-tests revealed widespread and significant alterations in AD/MCI compared to CN individuals. AD/MCI participants exhibited higher dwell times and increased self-transitions, indicating reduced neural flexibility and a tendency to remain in specific connectivity states. In contrast, CN individuals showed more diverse and recurrent transitions, reflecting greater adaptability. Secondary transitions revealed widespread selective switching in CN, whereas AD/MCI showed reduced cross-state engagement.A classification model trained on 6,960 dynamic features achieved strong performance in distinguishing AD/MCI from CN (mean AUC ≈ 0.85). These findings highlight the potential of guided dFNC as a biomarker framework for early-stage AD detection using resting-state fMRI.

Mild Behavioral Impairment as a Predictor of Functional Status.

BackgroundSleep–wake rhythms are critical for the development of Alzheimer’s disease (AD). However, the relationship of sleep disturbance, APOE ε4, and amyloid-β (Aβ) accumulation remains unclear. Thus, this study investigated the potential role of APOE ε4 allele in the association between sleep disturbance and brain Aβ burden among cognitively normal (CN) older adults.MethodsIn this cross-sectional study, data were obtained from the Alzheimer’s Disease Neuroimaging Initiative (ADNl) Database. The sample consisted of CN individuals aged between 55 and 90 years with Aβ positron emission tomography scan, APOE genotype, and sleep assessment using the Neuropsychiatric Inventory.ResultsThe study included 1,000 CN participants, including 134 individuals with sleep disturbances and 306 APOE ε4 carriers (APOE ε4+). After adjusting for sex, age, years of education, and marital status, sleep disturbance was not associated with a higher Aβ burden among participants. However, a significant interaction between sleep disturbance and APOE ε4 on regional standardized uptake value ratios was observed, such as in the left hippocampus. Subgroup analysis revealed that sleep disturbance could affect the AD-sensitive brain regions in the APOE ε4 + group. Furthermore, the subjective severity of sleep disturbance was linearly associated with a more significant Aβ brain burden in the APOE ε4 + group.ConclusionThis study demonstrated that CN individuals with both APOE ε4 + status and sleep disturbance exhibited greater Aβ burden. Understanding the relationship between sleep and Aβ in CN older adults may inform sleep interventions that could reduce early Aβ accumulation and delay the onset of cognitive dysfunction associated with early AD.

Effects of regional white matter hyperintensities and β-amyloid on domain-specific cognition and progression to dementia.

Background: Mild Cognitive Impairment (MCI) represents an intermediate stage between normal cognitive aging and Alzheimer's Disease (AD). Early and accurate identification of MCI is crucial for implementing interventions that may delay or prevent further cognitive decline. This study aims to develop a machine learning-based model for differentiating between Cognitively Normal (CN) individuals and MCI patients using data from the Alzheimer's Disease Neuroimaging Initiative (ADNI). Methods: An ensemble classification approach was designed by integrating Extra Trees, Random Forest, and Light Gradient Boosting Machine (LightGBM) algorithms. Feature selection emphasized clinically relevant biomarkers, including Amyloid-β 42, phosphorylated tau, diastolic blood pressure, age, and gender. The dataset was split into training and held-out test sets. A probability thresholding strategy was employed to flag uncertain predictions for potential deferral, enhancing model reliability in borderline cases. Results: The final ensemble model achieved an accuracy of 83.2%, a recall of 80.2%, and a precision of 86.3% on the independent test set. The probability thresholding mechanism flagged 23.3% of cases as uncertain, allowing the system to abstain from low-confidence predictions. This strategy improved clinical interpretability and minimized the risk of misclassification in ambiguous cases. Conclusions: The proposed AI-driven ensemble model demonstrates strong performance in classifying MCI versus CN individuals using multimodal ADNI data. Incorporating a deferral mechanism through uncertainty estimation further enhances the model's clinical utility. These findings support the integration of machine learning tools into early screening workflows for cognitive impairment.

BackgroundHandwriting and speech are served as reliable signatures for detecting cognitive decline, playing a pivotal role in the early diagnosing Alzheimer's disease (AD) and mild cognitive impairment (MCI). However, current unimodal approaches for diagnosing AD and MCI have demonstrated constraints in classification accuracy, potentially overlooking the synergistic value of combining handwriting and speech data.ObjectivePresenting an innovative multi-modal screening classification model, that harnesses handwriting and speech analysis to enhance MCI detection, aiming to overcome the constraints of single-modality approaches by integrating data from both modalities, thereby improving diagnostic accuracy.MethodsProposing a multimodal classification model based on gated recurrent unit (GRU) and attention mechanism, treating handwriting and speech data as sequence inputs. The model was constructed and tested on a dataset of 41 participants, including 20 MCI patients and 21 cognitively normal (CN) individuals. To mitigate the risk of overfitting due to the small sample size, we employed a 10-fold cross-validation strategy to ensure the robustness of the results.ResultsOur multimodal classification model achieved an accuracy of 95.2% for MCI versus CN individuals, which shows a significant improvement compared to the results of single-modality. This result indicates the effectiveness of the cross-fusion model in enhancing classification performance, offering a promising approach for the early diagnosis of neurodegenerative diseases.ConclusionsThe proposed GRU_CA effectively improves early MCI detection by fusing handwriting and speech data, outperforming a single modality. It shows strong potential for deployment in primary healthcare settings and establishes a foundation for future research on more complex diagnostic tasks, including CN, MCI, and AD classification, as well as longitudinal studies.

Purpose: To establish a normative database of choroidal vascularity index (CVI) in cognitively normal adults. Methods: Cognitively healthy volunteers who had a Montreal Cognitive Assessment score of 23 or higher were included. Exclusion criteria included diabetes mellitus, uncontrolled hypertension, glaucoma, optic neuropathy, vitreoretinal disorders, intraocular surgery other than cataract or refractive, and visual acuity worse than 20/40. Enhanced depth optical coherence tomography images were taken. Choroidal vascularity index was calculated from total choroidal area and luminal area. Results: Overall average CVI was 67.0% ± 0.026%, mean total choroidal area was 7.105 ± 4.618, and luminal area was 4.676 ± 2.843. There was no significant influence of sex on CVI (mean 66.9% females, 67.2% males, P-value 0.444). Age was not strongly associated with CVI with R-squared values all below 0.021. A total of 504 eyes (89.4%) were from White participants, 43 from Black participants (7.50%), six from Asian participants, and 12 from other/unreported race. Race did not have a significant influence on mean CVI (67.1% White, 67.3% Black, 66.5% Asian, 64.0% other/not reported, P = 0.073). No statistically significant associations were identified between CVI and the presence of hypertension or cardiac disease. Conclusion: Choroidal vascularity index is a durable metric across several demographic factors and presence of hypertension and/or cardiac disease. After additional studies, CVI may be a useful biomarker for neurologic, retinal, and choroidal diseases.

The sensitive Amnesia Light and Brief Assessment (ALBA) is a valid 3-min test of 4 tasks indicative of mild cognitive deficits.

Alzheimer's disease (AD), the leading cause of dementia, poses a growing global health challenge due to its rising prevalence and socioeconomic impact. Investigating metabolic alterations associated with white matter integrity (WMI) could provide critical insights into AD pathogenesis and identify potential therapeutic targets. This cross-sectional study explored the associations between amino acid (AA) profiles, assessed via ultra-high-performance liquid chromatography (UHPLC), and WMI metrics derived from diffusion tensor imaging (DTI) in individuals across the AD continuum. A total of 176 participants from the Alzheimer's Disease Neuroimaging Initiative (ADNI) were included and grouped into cognitively normal (CN) individuals (n = 54), patients with mild cognitive impairment (MCI, n = 88), and AD patients (n = 34). WMI was evaluated using DTI-derived metrics, including fractional anisotropy (FA), mean diffusivity (MD), radial diffusivity (RD), and axial diffusivity (AD). AA profiling was conducted using an appropriate panel. Regression analyses, adjusted for age, gender, and education, was used to identify significant associations between AA levels and WMI. Distinct AA alterations were associated with white matter microstructural integrity across study groups. In CN individuals, higher levels of arginine, glycine, and threonine correlated with decreased FA and increased MD, indicating reduced white matter integrity. Conversely, in AD patients, aspartate, glutamate, and histidine exhibited opposite associations, showing positive correlations with FA and negative correlations with MD, suggesting potential neuroprotective or compensatory mechanisms. These findings underscore the associations between AA patterns and white matter integrity and their potential role as AD progression markers. Further investigations into these AA metabolism pathways may identify novel biomarkers for early diagnosis and targets for therapeutic interventions.

Background/Objectives: Gamma oscillations (30-100 Hz), which are essential for memory, attention, and cortical synchronization, remain underexplored in Alzheimer's disease (AD) research. While resting-state EEG studies have predominantly examined lower frequency bands (delta to beta), gamma activity may more accurately reflect early synaptic dysfunction and other mechanisms relevant to AD pathophysiology. AD is a common age-related neurodegenerative disorder frequently associated with altered resting-state EEG (rEEG) patterns. This study analyzed gamma power spectral density (PSD) during eyes-open (EOR) and eyes-closed (ECR) resting-state EEG in AD patients compared to cognitively normal (CN) individuals. Methods: rEEG data from 534 participants (269 CN, 265 AD) aged 40-90 were analyzed. Quantitative EEG (qEEG) analysis focused on the gamma band (30-100 Hz) using PSD estimation with the Welch method, coherence matrices, and coherence-based functional connectivity. Data preprocessing and analysis were performed using EEGLAB and Brainstorm in MATLAB R2024b. Group comparisons were conducted using ANOVA for unadjusted models and linear regression with age adjustment using log10-transformed PSD values in Python (version 3.13.2, 2025). Results: AD patients exhibited significantly elevated gamma PSD in frontal and temporal regions during EOR and ECR states compared to CN. During ECR, gamma PSD was markedly higher in the AD group (Mean = 0.0860 ± 0.0590) than CN (Mean = 0.0042 ± 0.0010), with a large effect size (Cohen's d = 1.960, p < 0.001). Conversely, after adjusting for age, the group difference was no longer statistically significant (β = -0.0047, SE = 0.0054, p = 0.391), while age remained a significant predictor of gamma power (β = -0.0008, p = 0.019). Pairwise coherence matrix and coherence-based functional connectivity were increased in AD during ECR but decreased in EOR relative to CN. Conclusions: Gamma oscillatory activity in the 30-100 Hz range differed significantly between AD and CN individuals during resting-state EEG, particularly under ECR conditions. However, age-adjusted analyses revealed that these differences are not AD-specific, suggesting that gamma band changes may reflect aging-related processes more than disease effects. These findings contribute to the evolving understanding of gamma dynamics in dementia and support further investigation of gamma PSD as a potential, age-sensitive biomarker.

Aligning biomarker evidence with clinical presentation in early Alzheimer disease (AD) is essential for improving diagnosis, prognosis, and interventions. This study evaluates the relationship between cognitive impairment, future decline, and phosphorylated tau levels in plasma and CSF in predementia AD. This longitudinal observational study included predementia cases and controls from 2 independent cohorts: the Norwegian Dementia Disease Initiation (DDI) and Canadian Pre-Symptomatic Evaluation of Experimental or Novel Treatments for Alzheimer's Disease (PREVENT-AD). In DDI, cognitively normal (CN) and mild cognitive impairment (MCI) cases were classified using CSF Aβ42/40 ratio (A) and p-tau181 (T), whereas classification in PREVENT-AD (A) was based on amyloid PET scans. In DDI, we assessed CSF-plasma correlations for p-tau181, p-tau217, and p-tau231. Diagnostic accuracies were evaluated through receiver operating characteristic analyses. Linear mixed models evaluated p-tau associations with future memory decline. Between-group differences in plasma p-tau217 were assessed in both cohorts. In DDI (n = 431), participants were classified as CN A-/T- (n = 169), A+/T- (CN = 26, MCI = 24), A+/T+ (CN = 40, MCI = 105), and A-/T+ (CN = 34, MCI = 33), with a mean age of 64.1 years and 55.9% female. In PREVENT-AD (n = 190), participants were categorized as CN A- (n = 118), CN A+ (n = 49), and MCI A+ (n = 21), with a mean age of 67.8 years and 72.6% female. In DDI, plasma p-tau217 showed high accuracy in identifying A+ participants (areas under the curve [AUC]: 0.85) and a moderate correlation with CSF p-tau217 (rho = 0.65, p < 0.001). Diagnostic accuracy of plasma p-tau217 was greater in MCI A+ (AUC: 0.89) than in CN A+ (AUC: 0.79, p < 0.05) and in A+/T+ (AUC: 0.88) vs A+/T- (AUC: 0.78, p < 0.05). p-Tau181 and p-tau231 had weaker CSF-plasma correlations (rho = 0.47 and rho = 0.32, p < 0.001) and were less associated with cognitive status in A+ individuals. Higher plasma p-tau217 in A+ MCI vs A+ CN individuals (p < 0.001) was confirmed in PREVENT-AD. All CSF p-tau markers, but only plasma p-tau217, were associated with future memory decline (β = 0.05, p < 0.05). Our findings suggest that, unlike p-tau181 and p-tau231, plasma p-tau217 consistently aligns with cognitive status in A+ individuals and better reflects CSF biomarker abnormalities, reducing discrepancies between clinical and biochemical findings. Its association with baseline and future memory decline highlights its diagnostic and prognostic value, particularly when CSF analysis or PET is unavailable.

Recent developments in artificial intelligence (AI) have introduced new technologies that can aid in detecting cognitive decline. This study developed a voice-based AI model that screens for cognitive decline using only a short conversational voice sample. The process involved collecting voice samples, applying machine learning (ML), and confirming accuracy through test data. The AI model extracts multiple voice features from the collected voice data to detect potential signs of cognitive impairment. Data labeling for ML was based on Mini-Mental State Examination scores: scores of 23 or lower were labeled as “cognitively declined (CD),” while scores above 24 were labeled as “cognitively normal (CN).” A fully coupled neural network architecture was employed for deep learning, using voice samples from 263 patients. Twenty voice samples, each comprising a one-minute conversation, were used for accuracy evaluation. The developed AI model achieved an accuracy of 0.950 in discriminating between CD and CN individuals, with a sensitivity of 0.875, specificity of 1.000, and an average area under the curve of 0.990. This voice AI model shows promise as a cognitive screening tool accessible via mobile devices, requiring no specialized environments or equipment, and can help detect CD, offering individuals the opportunity to seek medical attention.

CSF α-synuclein aggregation is associated with APOE ε4 and progressive cognitive decline in Alzheimer's disease.

Cortical regions such as parietal area H (PH) and the fundus of the superior temporal sulcus (FST) are involved in higher visual function and may play a role in dementia with Lewy bodies (DLB), which is frequently associated with hallucinations. The authors evaluated functional connectivity between these two regions for distinguishing participants with DLB from those with Alzheimer's disease (AD) or mild cognitive impairment (MCI) and from cognitively normal (CN) individuals to identify a functional connectivity MRI signature for DLB. Eighteen DLB participants completed cognitive testing and functional MRI scans and were matched to AD or MCI and CN individuals whose data were obtained from the Alzheimer's Disease Neuroimaging Initiative database (https://adni.loni.usc.edu). Images were analyzed with data from Human Connectome Project (HCP) comparison individuals by using a machine learning-based subject-specific HCP atlas based on diffusion tractography. Bihemispheric functional connectivity of the PH to left FST regions was reduced in the DLB group compared with the AD and CN groups (mean±SD connectivity score=0.307±0.009 vs. 0.456±0.006 and 0.433±0.006, respectively). No significant differences were detected among the groups in connectivity within basal ganglia structures, and no significant correlations were observed between neuropsychological testing results and functional connectivity between the PH and FST regions. Performances on clock-drawing and number-cancelation tests were significantly and negatively correlated with connectivity between the right caudate nucleus and right substantia nigra for DLB participants but not for AD or CN participants. The functional connectivity between PH and FST regions is uniquely affected by DLB and may help distinguish this condition from AD.

Dementia is a neurodegenerative syndrome marked by the accumulation of disease-specific proteins and immune dysregulation, including autoimmune mechanisms involving autoantibodies. Current diagnostic methods are often invasive, time-consuming, or costly. This study explores the use of proteome-wide autoantibody screening (PWAbS) for noninvasive dementia diagnosis by analyzing serum samples from Alzheimer's disease (AD), dementia with Lewy bodies (DLB), and age-matched cognitively normal individuals (CNIs). Serum samples from 35 subjects were analyzed utilizing our original wet protein arrays displaying more than 13,000 human proteins. PWAbS revealed elevated gross autoantibody levels in AD and DLB patients compared to CNIs. A total of 229 autoantibodies were differentially elevated in AD and/or DLB, effectively distinguishing between patient groups. Machine learning models showed high accuracy in classifying AD, DLB, and CNIs. Gene ontology analysis highlighted autoantibodies targeting neuroactive ligands/receptors in AD and lipid metabolism proteins in DLB. Notably, autoantibodies targeting neuropeptide B (NPB) and adhesion G protein-coupled receptor F5 (ADGRF5) showed significant correlations with clinical traits including Mini Mental State Examination scores. The study demonstrates the potential of PWAbS and artificial intelligence integration as a noninvasive diagnostic tool for dementia, uncovering biomarkers that could enhance understanding of disease mechanisms. Limitations include demographic differences, small sample size, and lack of external validation. Future research should involve longitudinal observation in larger, diverse cohorts and functional studies to clarify autoantibodies' roles in dementia pathogenesis and their diagnostic and therapeutic potential.

Brain iron accumulation is recognized as a cause and therapeutic target in Alzheimer's disease (AD). We investigated the differences in both volume and iron accumulation between cognitively normal (CN) older adults and patients with amnestic mild cognitive impairment (aMCI). Additionally, we assessed which combination of these measures best explains the group differences in visual and verbal memory performance. We retrospectively analyzed data from 48 patients with aMCI and 33 age-matched CN individuals. Structural differences were investigated using voxel-based comparisons of T1-weighted magnetic resonance images. Differences in iron accumulation were investigated using voxel-based comparisons of quantitative susceptibility mapping (QSM) images. Subsequently, significant clusters from these voxel-based analyses (amygdala, posterior cingulate cortex, precuneus, lateral occipital cortex, and pericalcarine cortex) were entered into a stepwise regression to predict verbal and visual memory scores, while accounting for age, sex, and education as covariates. In comparison to CN, patients with aMCI had significantly lower scores in both verbal and visual memory tests (p < 0.001). The T1-weighted voxel-based morphometry (VBM) results showed significant hippocampal atrophy in the aMCI group relative to CN individuals. The QSM-VBM results showed increased iron accumulation in the amygdala, posterior cingulate cortex, precuneus, lateral occipital cortex, and pericalcarine cortex (FWE-corrected p < 0.05). Lower hippocampal volume (B = 2015.91, SE = 469.61, p < 0.001) and higher posterior cingulate cortex susceptibility (B = -189.63 SE = 89.11, p = 0.037) were significant predictors of verbal memory. For visual memory, higher lateral occipital susceptibility (B = -659. 96, SE = 253.03, p = 0.011) was significant imaging predictor. These results suggest that iron accumulates in regions where atrophy has not yet occurred, suggesting that iron may serve as an earlier imaging marker of neurodegeneration compared to volume atrophy. Further studies are needed to investigate the longitudinal relationship between brain volume and iron accumulation during cognitive decline.

A network neuroscience perspective can significantly advance the understanding of neurodegenerative disorders, particularly frontotemporal dementia (FTD). This study employed an innovative multiplex connectomics approach, integrating cortical thickness (CTH) and fluorodeoxyglucose-positron emission tomography (FDG-PET) in a dual-layer model to investigate network alterations in FTD subtypes across two geographically distinct sites. The cohort included groups of behavioral variant FTD (bvFTD), primary progressive aphasia (PPA), mild cognitive impairment (MCI), and cognitively normal (CN) individuals who were analyzed from two separate sites. Site 1 included 28 bvFTD, 20 PPA, and 27 MCI participants, whereas Site 2 included 26 bvFTD, 43 PPA, and 43 CN individuals, respectively. Utilizing CTH and FDG-PET data after standard preprocessing, a multiplex network pipeline in BRAPH2 toolbox was used to derive multiplex participation coefficient (MPC) between the groups. The analysis revealed an increase in global MPC as an indicator of disease in PPA at both sites. Additionally, nodal MPC alterations in the anterior cingulate, frontal, and temporal lobes in PPA were compared with bvFTD. Comparisons with the CN showed that nodal MPC alterations were more extensive in PPA when compared with bvFTD. These findings underscore the potential utility of multiplex connectomes for identifying network disruptions in neurodegenerative disorders, offering promising implications for future research and clinical applications.

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

Cognitive impairment is a common nonmotor symptom among individuals with Parkinson's disease (PD). Although cognitive impairment generally develops progressively, individuals with PD-associated mild cognitive impairment (PD-MCI) may revert to being cognitively normal (CN), which is referred to as PD-MCI reversion. Previous studies are inconsistent in whether PD-MCI reverters are at greater risk for PD-MCI recurrence relative to CN individuals. Even less is known about how PD-MCI reverters compare with CN individuals or PD-MCI nonreverters in terms of neurodegenerative biomarkers. The authors examined group differences (CN, PD-MCI reversion, and PD-MCI nonreversion) in cerebrospinal fluid (CSF) markers of Alzheimer's disease (AD), including amyloid beta, tau (total [t-tau] and phosphorylated [p-tau]), and alpha-synuclein. Data from the longitudinal international multisite Parkinson Progression Marker Initiative were used. Participants were newly diagnosed as having PD (N=430) and completed a battery of neurocognitive assessments at baseline and annual follow-ups for up to 5 years. Participants were classified as CN, PD-MCI reverters, or PD-MCI nonreverters on the basis of the first two neurocognitive assessments. The PD-MCI nonreversion group had greater p-tau:amyloid beta and t-tau:amyloid beta ratios relative to the PD-MCI reversion group. The CN and PD-MCI reversion groups did not significantly differ in any of the CSF markers. PD-MCI reverters may have a more favorable trajectory in terms of AD pathology relative to PD-MCI nonreverters. Future studies are needed to verify whether PD-MCI reversion may represent an intermediate prognostic group between CN individuals and those with MCI nonreversion.