The classification of Alzheimer’s disease (AD) using deep learning methods has shown promising results, but successful application in clinical settings requires a combination of high accuracy, short processing time, and generalizability to various populations. In this study, we developed a convolutional neural network (CNN)-based AD classification algorithm using magnetic resonance imaging (MRI) scans from AD patients and age/gender-matched cognitively normal controls from two populations that differ in ethnicity and education level. These populations come from the Seoul National University Bundang Hospital (SNUBH) and Alzheimer’s Disease Neuroimaging Initiative (ADNI). For each population, we trained CNNs on five subsets using coronal slices of T1-weighted images that cover the medial temporal lobe. We evaluated the models on validation subsets from both the same population (within-dataset validation) and other population (between-dataset validation). Our models achieved average areas under the curves of 0.91–0.94 for within-dataset validation and 0.88–0.89 for between-dataset validation. The mean processing time per person was 23–24 s. The within-dataset and between-dataset performances were comparable between the ADNI-derived and SNUBH-derived models. These results demonstrate the generalizability of our models to different patients with different ethnicities and education levels, as well as their potential for deployment as fast and accurate diagnostic support tools for AD.

Background Despite strong evidence that emotional support has a protective effect on cognitive decline, the neural basis for how an individual’s emotional support is associated with cognition is unknown. We investigated if the hippocampus mediates the relationship between emotional support and cognition in older adults. Method Four hundred and ten nondemented Korean older adults were classified into two groups according to their Medical Outcomes Study–Social Support Survey emotional support scores: the poor emotional support (score ≤ 25th percentile of entire sample) and normal emotional support (score > 25th percentile of entire sample) groups. Cognitive function was assessed using the Verbal Memory Score and Consortium to Establish a Registry for Alzheimer’s Disease Assessment Packet Neuropsychological Assessment Battery total score (CERAD-TS). Left and right hippocampal volume were obtained using 3T T1-weighted magnetic resonance images. Mediation analyses were conducted. Results In the association of emotional support with Verbal Memory Score, left hippocampal volume played a mediating role (indirect effect = 0.40, SE = 0.21, bias-corrected 95% confidence interval = 0.04, 0.86). In the association of emotional support with CERAD-TS, both left (indirect effect = 0.82, SE = 0.45, bias-corrected 95% confidence interval = 0.09, 1.83) and right (indirect effect = 0.51, SE = 0.32, bias-corrected 95% confidence interval = 0.02, 1.24) HPVs played mediating roles. Conclusions The hippocampus may mediate the association between emotional support and cognition. Higher levels of emotional support may be associated with better verbal memory and global cognition via larger HPV.

Background To investigate the association between pineal gland volume and symptoms of rapid eye movement (REM) sleep behavior disorder (RBD) in Alzheimer’s disease (AD) patients without any feature of dementia with Lewy bodies. Methods We enrolled 296 community-dwelling probable AD patients who did not meet the diagnostic criteria for possible or probable dementia with Lewy bodies. Among them, 93 were amyloid beta (Aβ) positive on 18F-florbetaben amyloid brain positron emission tomography. We measured RBD symptoms using the REM Sleep Behavior Disorder Screening Questionnaire (RBDSQ) and defined probable RBD (pRBD) as the RBDSQ of 5 or higher. We manually segmented pineal gland on 3T structural T1-weighted brain magnetic resonance imaging. Results The participants with pRBD had smaller pineal parenchyma volume (VPP) than those without pRBD (p <  0.001). The smaller the VPP, the more severe the RBD symptoms (p <  0.001). VPP was inversely associated with risk of prevalent pRBD (odds ratio = 0.909, 95% confidence interval [CI] = 0.878–0.942, p <  0.001). Area under the receiver operator characteristic curve for pRBD of VPP was 0.80 (95% CI = 0.750–0.844, p <  0.0001). These results were not changed when we analyzed the 93 participants with Aβ-positive AD separately. Conclusions In AD patients, reduced pineal gland volume may be associated with RBD.

Highlights We constructed WPM from healthy elderly Koreans. WPM may serve as a tool to study pathology and normal aging of distribution of WMH. WPM provides a prominent atlas of the age related distribution of WMH.

<b><i>Introduction:</i></b> The irregular shapes of white matter hyperintensities (WMHs) are associated with poor cognitive function, diabetes, or lacunes. However, the association between the WMH shape and dementia remains understudied. We investigated the association between the calculated shape index of WMH and the diagnosis of dementia and cognitive function. <b><i>Methods:</i></b> The inverse sphericity index (ISI_WMH) and volume of WMHs (VOL_WMH) were compared among 82 participants with normal cognition, 82 with Alzheimer’s dementia (AD), and 82 with subcortical vascular dementia (SVD). We examined the associations of ISI_WMH and VOL_WMH with the modified Hachinski Ischemic Score (mHIS), diagnosis of AD and SVD, and cognitive test scores, using linear, multinomial, or hierarchical linear regression models. <b><i>Results:</i></b> The mHIS was associated with both ISI_WMH (β = 0.326, <i>p</i> &#x3c; 0.001) and VOL_WMH (β = 0.299, <i>p</i> &#x3c; 0.001). Both ISI_WMH and VOL_WMH were associated with the SVD diagnosis (odds ratio [OR] = 2.685, <i>p</i> = 0.002, ISI_WMH; OR = 2.597, <i>p</i> = 0.005, VOL_WMH), but not with AD. The SVD diagnosis was better explained when the multinomial regression model included both ISI_WMH and VOL_WMH instead of VOL_WMH alone (χ² = 20.768, df = 2, <i>p</i> &#x3c; 0.001). The Trail Making Test-D (TMT-D) scores of the SVD patients were associated with both ISI_WMH (β = 0.308) and VOL_WMH (β = 0.293). <b><i>Conclusion:</i></b> An irregular WMH shape may be associated with the high cerebrovascular component of cognitive impairment and the diagnosis and low cognitive flexibility of SVD, which may improve the prediction of SVD diagnosis when used in combination with WMH volume.