In Alzheimer's continuum (a comprehensive of preclinical Alzheimer's disease [AD], mild cognitive impairment [MCI] due to AD, and AD dementia), cognitive dysfunctions are often related to cortical atrophy in specific brain regions. The purpose of this study was to investigate the association between anatomical pattern of cortical atrophy and specific neuropsychological deficits. A total of 249 participants with Alzheimer's continuum (125 AD dementia, 103 MCI due to AD, and 21 preclinical AD) who were confirmed to be positive for amyloid deposits were collected from the memory disorder clinic in the department of neurology at Samsung Medical Center in Korea between September 2013 and March 2018. To analyze neuropsychological test-specific neural correlates representing the relationship between cortical atrophy measured by cortical thickness and performance in specific neuropsychological tests, a linear regression analysis was performed. Two neural correlates acquired by 2 different standardized scores in neuropsychological tests were also compared. Cortical atrophy in several specific brain regions was associated with most neuropsychological deficits, including digit span backward, naming, drawing-copying, verbal and visual recall, semantic fluency, phonemic fluency, and response inhibition. There were a few differences between 2 neural correlates obtained by different z-scores. The poor performance of most neuropsychological tests is closely related to cortical thinning in specific brain areas in Alzheimer's continuum. Therefore, the brain atrophy pattern in patients with Alzheimer's continuum can be predict by an accurate analysis of neuropsychological tests in clinical practice.

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.

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. 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. 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. 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.

Clinical trials for Alzheimer’s disease (AD) face multiple challenges, such as the high screen failure rate and the even allocation of heterogeneous participants. Artificial intelligence (AI), which has become a potent tool of modern science with the expansion in the volume, variety, and velocity of biological data, offers promising potential to address these issues in AD clinical trials. In this review, we introduce the current status of AD clinical trials and the topic of machine learning. Then, a comprehensive review is focused on the potential applications of AI in the steps of AD clinical trials, including the prediction of protein and MRI AD biomarkers in the prescreening process during eligibility assessment and the likelihood stratification of AD subjects into rapid and slow progressors in randomization. Finally, this review provides challenges, developments, and the future outlook on the integration of AI into AD clinical trials.