Alzheimer''s and Dementia: Diagnosis, Assessment and Disease Monitoring最新文献

Introduction: Alzheimer's disease (AD) is a progressive neurodegenerative disorder that needs better predictive tools. Using the National Alzheimer's Coordinating Center Uniform Data Set, this study developed machine learning (ML) models and a practical clinical tool for AD prediction.

Methods: Data from 52,537 individuals (22,371 with AD) and more than 200 variables were processed with MissForest imputation and genetic algorithm-based selection. Multiple ML models were trained, and interpretability was performed using SHAP and permutation importance. A LightGBM model was refined through iterative backward feature elimination (IBFE) followed by manual refinement.

Results: LightGBM performed best (receiver operating characteristic-area under the curve [ROC-AUC] 0.91, accuracy 82.0%). Key predictors included arthritis, age, body mass index, and heart rate. A 19-feature model retained accuracy (81.2%) and ROC-AUC (0.90).

Discussion: This lightweight tool predicts AD using mostly routine variables. Limitations include its cross-sectional nature, and would need external validation. An interactive web app and GitHub resource are available.

Highlights: Developed a lightweight ML based tool using 19 routinely available features.The lightweight model achieved an ROC-AUC of 0.90 for Alzheimer's disease prediction on NACC multicenter data.Genetic algorithm, IBFE, and manual refinement enabled optimal feature selection.Tool hosted on an open-access platform for clinical and research use.SHAP analysis provided model interpretability and feature-level insights.

Introduction: Dual cognitive-motor deficit in aging is a strong predictor of dementia, yet its effects on vulnerable gray matter region microstructure remain unexplored.

Methods: This study classified 582 individuals aged 36 to 90 into cognitive-motor deficit, isolated cognitive or motor deficit, and control groups. Microstructural differences in 27 temporal and motor-related gray matter (GM) regions and white matter (WM) tracts were assessed using diffusion tensor imaging and mean apparent propagator, a technique well suited for GM analysis.

Results: We found widespread microstructural alterations among individuals with dual cognitive-motor deficit. These changes were not observed in isolated cognitive or motor deficits after multiple comparisons correction.

Discussion: Dual cognitive-motor deficit is associated with distinct microstructural features that are hypothesized to indicate reduced cellular density in temporal GM, decreased fiber coherence, and potential demyelination in WM tracts. These findings could help understand brain aging and facilitate interventions to slow neurodegeneration and delay dementia onset.

Highlights: Dual cognitive-motor deficit strongly predicts dementia in older adults.Five hundred eighty-two individuals were classified by cognitive and motor deficiency.Mean apparent propagator magnetic resonance imaging (MRI) and diffusion tensor imaging identified widespread microstructural brain alterations.Only the dual deficit showed significant gray matter and white matter differences after correction.Results support early detection of dementia via diffusion MRI microstructure metrics.

Introduction: We compared the accuracy of pattern-recognition protocols to prospectively identify Alzheimer's disease and related dementias (ADRD) and differentiate these from normal aging or stroke.

Methods: Patterns of cognitive decline in cognitively unimpaired participants who completed ≥ 5 assessments for the Health and Retirement Study were examined to identify dementia/stroke and compared to both recorded clinical and objective diagnoses of amnestic cognitive impairment (aCI) and dementia. We report prevalence and sensitivity/specificity to detect new-onset ADRD and stroke.

Results: ADRD-related accelerated cognitive decline was identified in 372 (14.6%) participants, while stepwise decline consistent with stroke was identified in 917 (36.1%) participants. Accelerated decline was found preceding 75.8%/76.7% cases of aCI and objective dementia, respectively. Sensitivity for accelerated decline to detect aCI/objective dementia was excellent (96.2%/91.9%). Stepwise decline preceded diagnosis with executive cognitive impairment (eCI)/clinical stroke in 40.0%/43.3% of participants, and sensitivity was moderate for eCI/clinical stroke (45.3%/58.8%).

Discussion: Longitudinal patterns of cognitive decline can help differentially diagnose ADRD from stroke in longitudinal studies of cognitive decline.

Highlights: Pattern recognition identified 95.3% of all cases of dementia in this study.Sensitivity of accelerated cognitive decline to detect incident dementia was 94.3%.Differential diagnosis for dementia might begin to rely on longitudinal cognition.Pattern recognition worked in cases of clinically and algorithmically diagnosed dementia.

Background: Previous studies on retinal changes in Alzheimer's disease (AD) using optical coherence tomography (OCT) and electroretinography (ERG) based on clinical diagnostic criteria have yielded inconsistent results. We evaluated retinal structure and function in subjects classified using clinical and biological definitions.

Methods: Fifty-nine included subjects underwent comprehensive neuropsychiatric and ophthalmic evaluations, including OCT and ERG with photopic negative response (PhNR). Amyloid status was determined by ¹¹C-Pittsburgh compound B positron emission tomography (PET).

Results: No significant differences in evaluated OCT and ERG parameters were found between cognitively impaired and unimpaired groups. Amyloid-positive subjects showed significantly thinner macular, inner plexiform, and inner nuclear layers, plus ERG abnormalities (reduced PhNR, smaller waves amplitudes, prolonged a-wave latency) (p < 0.05). ERG outperformed OCT in discriminating amyloid status (area under the curve [AUC] = 0.84). Standardized uptake value ratio (SUVr) correlated with Mini-Mental State Examination (MMSE; r = 0.62, p < 0.05).

Discussion: Biomarker-based classification revealed retinal changes, affecting both inner retina and photoreceptors, not detected using clinical criteria.

Highlights: Retinal studies in Alzheimer's disease yield mixed results when based on clinical criteria.Amyloid positron emission tomography classification permits recognition of retinal changes missed clinically.Optical coherence tomography (OCT) shows early macular thinning in amyloid beta positive subjects at the expense of inner layers.Electroretinography detects outer retinal dysfunction, indicating broader involvement.Retinal function loss on electroretinography precedes inner/outer changes on OCT.

Background: Primary care offers ideal opportunities for early detection of cognitive impairment, yet clinics face significant barriers to routine screening. MyCog is an electronic health record-integrated tablet application self-administered during a primary care visit designed to overcome barriers to screening.

Methods: We compared MyCog performance between 65 adults age 65+ with diagnosed cognitive impairment and 80 cognitively normal adults. Five modeling approaches achieved consensus to select consistently discriminative variables for the final detection algorithm. Performance was primarily assessed via receiver operating characteristic area under the curve (AUC), sensitivity, specificity, and accuracy.

Results: All models demonstrated strong diagnostic performance (AUC 0.817 to 0.873). Memory accuracy and executive function efficiency scores were consistently selected as predictors of impairment across models. The final logistic regression achieved AUC 0.890, with sensitivity 0.723 to 0.831, specificity 0.788 to 0.912, and accuracy 0.807 to 0.828 depending on threshold.

Discussion: Findings suggest MyCog accurately detects cognitive impairment via a streamlined self-administered app that efficiently fits into primary care workflows.

Introduction: Differentiating dementia with Lewy bodies (DLB) from Alzheimer's disease (AD) is challenging. Seed amplification assay (SAA) is sensitive for the detection of misfolded α-synuclein.

Methods: Patients with DLB (N = 31) and AD (N = 25) were recruited and evaluated. Misfolded α-synuclein was assessed in cerebrospinal fluid (CSF), skin, urine, and olfactory mucosa using SAA.

Results: The accuracy of α-synuclein-SAA for DLB was 87% (95% confidence interval [CI]: 77% to 98%) in CSF, 85% (95% CI: 75% to 98%) in skin, 58% (95% CI: 47% to 69%) in olfactory mucosa, and 59% (95% CI: 51% to 66%) in urine. The core symptoms - fluctuations, REM sleep behavior disorder, and parkinsonism - had accuracies for SAA positivity of ≥79%. Notably, 95% of SAA-positive patients also had hyposmia.

Discussion: These findings support the use of CSF and skin α-synuclein-SAAs as diagnostic tools for DLB, with strong associations between SAA and clinical phenotype. In particular, intact olfactory function is associated with a lower risk of SAA positivity.

Highlights: CSF and skin biopsies show high diagnostic accuracy for α-synuclein, demonstrating good concordance.Strong correlations exist between core symptoms of DLB and pathological α-synuclein.A very high sensitivity of hyposmia for pathological α-synuclein is observed.A novel proof-of-concept is offered for the potential detection of pathological α-synuclein in urine, marking the first such comparative analysis between patients with DLB and AD.

Introduction: This survey investigated perspectives of research and clinical professionals on optimal content and features of measurement of self-perceived cognitive functioning.

Methods: Respondents were professionals working with older adults with self-reported cognitive concerns. The survey addressed views on harmonization and preferences for items, response formatting, practical features, and instrument validation. We evaluated item preferences in consideration of a previous statistical harmonization.

Results: Ninety professionals from 20 different countries completed the survey. Most professionals (87%) indicated a need for a harmonized instrument. Respondents agreed that an instrument should measure current ability alongside change therein, focus on memory, and adopt Likert scale responses. Recommendations for assessment timeframe, practical features, and validation priorities varied. Respondents differentially endorsed items previously found to be statistically informative.

Discussion: Respondents agreed on overarching measurement topics, with varying recommendations for specific content and features. Together with statistical information, these results provide a starting point for a harmonized instrument.

Highlights: Professionals see a need for a harmonized tool to measure cognitive concerns.Professionals have diverse preferences for measurement content and its validation.Item relevance as seen by professionals aligned considerably with statistical value.Integration of statistical information with expert and patient opinion is crucial.

Introduction: To address uncertainty about long-term clinical and economic impacts of an accurate dementia diagnosis, we evaluated the cost-effectiveness of adding amyloid positron emission tomography (PET) to memory clinic workup over 5 years.

Methods: Inverse probability weighting was used to balance covariates between PET (n = 440) and no-PET (n = 460) participants from the Amsterdam Dementia Cohort. Time in community following diagnosis, time alive, and costs were combined in cost-effectiveness analyses.

Results: PET participants lived longer in community (0.26 years, 95% confidence interval [CI]: 0.05 to 0.45) and overall (0.15, CI: 0.02 to 0.27), but did not have statistically different health insurance (€703, CI: -3974 to 5045) or total costs including institutionalization (-€8258, CI: -20,622 to 3377). The probability that PET was cost-effective for extending time in community was 76% at a €2530 willingness-to-pay threshold. The probability that PET yielded cost savings and was more effective for extending time alive was 90%.

Discussion: Findings in this observational cohort suggest that using amyloid PET in memory clinics may be cost-effective.

Highlights: Participants with an amyloid PET in a memory clinic work-up were compared to those without.The amyloid PET group spent more time in community and alive over 5 years of follow-up.Amyloid PET had a 76% chance to cost-effectively extend time in community in uncertainty analysis.