Search

Amyloid positivity is an early indicator of Alzheimer's disease and is necessary to determine the disease. In this study, a deep generative model is utilized to predict the amyloid positivity of cognitively normal individuals using proxy measures, such as structural MRI scans, demographic variables, and cognitive scores, instead of invasive direct measurements. Through its remarkable efficacy in handling imperfect datasets caused by missing data or labels, and imbalanced classes, the model outperforms previous studies and widely used machine learning approaches with an AUROC of 0.8609. Furthermore, this study illuminates the model's adaptability to diverse clinical scenarios, even when feature sets or diagnostic criteria differ from the training data. We identify the brain regions and variables that contribute most to classification, including the lateral occipital lobes, posterior temporal lobe, and APOE ϵ4 allele. Taking advantage of deep generative models, our approach can not only provide inexpensive, non-invasive, and accurate diagnostics for preclinical Alzheimer's disease, but also meet real-world requirements for clinical translation of a deep learning model, including transferability and interpretability., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2023 The Authors. Published by Elsevier B.V. All rights reserved.)

Objective: We investigated the frequency of β-amyloid (Aβ) positivity in 9 groups classified according to a combination of 3 different cognition states and 3 distinct levels of white matter hyperintensities (WMH) (minimal, moderate, and severe) and aimed to determine which factors were associated with Aβ after controlling for WMH and vice versa., Methods: A total of 1,047 individuals with subjective cognitive decline (SCD, n = 294), mild cognitive impairment (MCI, n = 237), or dementia (n = 516) who underwent Aβ PET scans were recruited from the memory clinic at Samsung Medical Center in Seoul, Korea. We investigated the following: (1) Aβ positivity in the 9 groups, (2) the relationship between Aβ positivity and WMH severity, and (3) clinical and genetic factors independently associated with Aβ or WMH., Results: Aβ positivity increased as the severity of cognitive impairment increased (SCD [15.7%], MCI [43.5%], and dementia [76.2%]), whereas it decreased as the severity of WMH increased (minimal [54.5%], moderate [53.9%], and severe [41.0%]) or the number of lacunes (0 [59.0%], 1-3 [42.0%], and >3 [23.4%]) increased. Aβ positivity was associated with higher education, absence of diabetes, and presence of APOE ε4 after controlling for cognitive and WMH status., Conclusion: Our analysis of Aβ positivity involving a large sample classified according to the stratified cognitive states and WMH severity indicates that Alzheimer and cerebral small vessel diseases lie on a continuum. Our results offer clinicians insightful information about the association among Aβ, WMH, and cognition., (© 2021 American Academy of Neurology.)

Background: Amyloid-β (Aβ) evaluation in amnestic mild cognitive impairment (aMCI) patients is important for predicting conversion to Alzheimer's disease. However, Aβ evaluation through Aβ positron emission tomography (PET) is limited due to high cost and safety issues., Objective: We therefore aimed to develop and validate prediction models of Aβ positivity for aMCI using optimal interpretable machine learning (ML) approaches utilizing multimodal markers., Methods: We recruited 529 aMCI patients from multiple centers who underwent Aβ PET. We trained ML algorithms using a training cohort (324 aMCI from Samsung medical center) with two-phase modelling: model 1 included age, gender, education, diabetes, hypertension, apolipoprotein E genotype, and neuropsychological test scores; model 2 included the same variables as model 1 with additional MRI features. We used four-fold cross-validation during the modelling and evaluated the models on an external validation cohort (187 aMCI from the other centers)., Results: Model 1 showed good accuracy (area under the receiver operating characteristic curve [AUROC] 0.837) in cross-validation, and fair accuracy (AUROC 0.765) in external validation. Model 2 led to improvement in the prediction performance with good accuracy (AUROC 0.892) in cross validation compared to model 1. Apolipoprotein E genotype, delayed recall task scores, and interaction between cortical thickness in the temporal region and hippocampal volume were the most important predictors of Aβ positivity., Conclusion: Our results suggest that ML models are effective in predicting Aβ positivity at the individual level and could help the biomarker-guided diagnosis of prodromal AD.

Background: Computerized versions of cognitive screening test could have advantages over pencil-and-paper versions by eliminating rater-dependent factors and saving the time required to score the tests and report the results. We developed a computerized cognitive screening test (Inbrain Cognitive Screening Test [Inbrain CST]) that takes about 30 minutes to administer on a touchscreen computer and is composed of neuropsychological tests already shown to be sensitive in detecting early cognitive decline in Alzheimer's disease (AD). The aims of this study were to 1) introduce normative data for Inbrain CST, 2) verify its reliability and validity, 3) assess clinical usefulness, and 4) identify neuroanatomical correlates of Inbrain CST., Methods: The Inbrain CST runs on the Microsoft Windows 10 operating system and comprises 7 subtests that encompass 5 cognitive domains: attention, language, visuospatial, memory, and executive functions. First, we recruited 480 cognitively normal elderly people (age 50-90) from communities nationwide to establish normative data for Inbrain CST. Second, we enrolled 97 patients from our dementia clinic (26 with subjective cognitive decline [SCD], 42 with amnestic mild cognitive impairment [aMCI], and 29 with dementia due to AD) and investigated sensitivity and specificity of Inbrain CST for discriminating cognitively impaired patients from those with SCD using receiver operating characteristic (ROC) curve analyses. Third, we compared the Inbrain CST scores with those from another neuropsychological test battery to obtain concurrent validity and assessed test-retest reliability. Finally, magnetic resonance imaging (MRI)-based cortical thickness analyses were performed to provide anatomical substrates for performances on the Inbrain CST., Results: First, in the normative sample, the total score on the Inbrain CST was significantly affected by age, years of education, and gender. Second, Inbrain CST scores among the three patient groups decreased in the order of SCD, aMCI, and AD dementia, and the ROC curve analysis revealed that Inbrain CST had good discriminative power for differentiating cognitively impaired patients from those with SCD. Third, the Inbrain CST subtests had high concurrent validity and test-retest reliability. Finally, in the cortical thickness analysis, each cognitive domain score and the total score of Inbrain CST showed distinct patterns of anatomical correlates that fit into the previously known brain-behavior relationship., Conclusion: Inbrain CST had good validity, reliability, and clinical usefulness in detecting cognitive impairment in the elderly. Furthermore, it showed neuroanatomical validity through MRI cortical thinning patterns. These results suggest that Inbrain CST is a useful cognitive screening tool with efficiency and validity to detect mild impairments in cognition in clinical settings., Competing Interests: Juoh Yun, Jongha Park, Jiho Yeom, and Dae-Seock Shin are employed at MIDAS IT, the company that produces the Inbrain CST used in this study. The company made no influence in collection and analysis of data. The other authors have no potential conflicts of interest to disclose., (© 2020 The Korean Academy of Medical Sciences.)

Recent improvements in neuroimaging and molecular markers of Alzheimer's disease (AD) have aided diagnosis in the early stage of the disease, which greatly increases the chance for successful prevention and treatment. However, the expanding resources for AD diagnosis are unlikely to benefit all elderly due to economic burden. Here, we aimed to develop an inexpensive and sensitive method to detect early-stage AD. A scenario for real-world social event memory test (SEMT) was created and filmed in 360° video. Participants watched the 7-min video through head-mounted display (HMD) and then answered questionnaire about the video. We categorized the SEMT score into recall, recognition, and place-matching scores and compared them to scores on the Mini-Mental State Examination and Seoul Verbal Learning Test. Using the SEMT scores, we built a logistic regression model that discriminated between amyloid positivity and negativity of the participants, with a cross-validation AUC. Furthermore, a classifier was created using support vector machine, which produced 93.8-95.1% sensitivity in classifying individuals into four groups of normal, mild cognitive impairment with or without amyloid, and AD elderly. The high correlation between the SEMT score and amyloid positivity in individuals who experienced virtual social gathering through an HMD opens a new possibility for early diagnosis of AD.

ObjectiveTo investigate the association between APOE genotype and β-amyloid (Aβ) burden, as measured by PET in patients with subcortical vascular cognitive impairment (SVCI) and those with Alzheimer disease-related cognitive impairment (ADCI).MethodsThis was a cross-sectional study of 310 patients with SVCI and 999 with ADCI. To evaluate the effects of APOE genotype or diagnostic group on Aβ positivity, we performed multivariate logistic regression analyses. Further distinctive underlying features of latent subgroups were examined by employing a latent class cluster analysis approach.ResultsIn comparison with ε3 homozygotes, in the ADCI group, ε2 carriers showed a lower frequency of Aβ positivity (odds ratio [OR] 0.43, 95% confidence interval [CI] 0.23-0.79), while in the SVCI group, ε2 carriers showed a higher frequency of Aβ positivity (OR 2.26, 95% CI 1.02-5.01). In particular, we observed an interaction effect of ε2 carrier status and diagnostic group on Aβ positivity (OR 5.12, 95% CI 1.93-13.56), in that relative to ε3 homozygotes, there were more Aβ-positive ε2 carriers in the SVCI group than in the ADCI group. We also identified latent subgroups of Aβ-positive APOE ε2 carriers with SVCI and Aβ-positive APOE ε4 carriers with ADCI.ConclusionsOur findings suggest that APOE ε2 is distinctly associated with Aβ deposition in patients with SVCI and those with ADCI. Our findings further suggest that there is a distinctive subgroup of Aβ-positive APOE ε2 carriers with SVCI among patients with cognitive impairment.