Background: Neuropsychological tests (NPTs) are important tools for informing diagnoses of cognitive impairment (CI). However, interpreting NPTs requires specialists and is thus time-consuming. To streamline the application of NPTs in clinical settings, we developed and evaluated the accuracy of a machine learning algorithm using multi-center NPT data. Methods: Multi-center data were obtained from 14,926 formal neuropsychological assessments (Seoul Neuropsychological Screening Battery), which were classified into normal cognition (NC), mild cognitive impairment (MCI) and Alzheimer’s disease dementia (ADD). We trained a machine learning model with artificial neural network algorithm using TensorFlow (https://www.tensorflow.org) to distinguish cognitive state with the 46-variable data and measured prediction accuracies from 10 randomly selected datasets. The features of the NPT were listed in order of their contribution to the outcome using Recursive Feature Elimination. Results: The ten times mean accuracies of identifying CI (MCI and ADD) achieved by 96.66 ± 0.52% of the balanced dataset and 97.23 ± 0.32% of the clinic-based dataset, and the accuracies for predicting cognitive states (NC, MCI or ADD) were 95.49 ± 0.53 and 96.34 ± 1.03%. The sensitivity to the detection CI and MCI in the balanced dataset were 96.0 and 96.0%, and the specificity were 96.8 and 97.4%, respectively. The ‘time orientation’ and ‘3-word recall’ score of MMSE were highly ranked features in predicting CI and cognitive state. The twelve features reduced from 46 variable of NPTs with age and education had contributed to more than 90% accuracy in predicting cognitive impairment. Conclusions: The machine learning algorithm for NPTs has suggested potential use as a reference in differentiating cognitive impairment in the clinical setting.

We aimed to present the study design of an independent validation cohort from the Korean Brain Aging Study for the Early Diagnosis and Prediction of Alzheimer’s disease (AD) (KBASE-V) and to investigate the baseline characteristics of the participants according to the AD clinical spectrum. We recruited 71 cognitively normal (CN) participants, 96 with subjective cognitive decline (SCD), 72 with mild cognitive impairment (MCI), and 56 with AD dementia (ADD). The participants are followed for three years. The Consortium to Establish a Registry for AD scores was significantly different between all of the groups. The logical memory delayed recall scores were significantly different between all groups, except between the MCI and ADD groups. The Mini-Mental State Examination score, hippocampal volume, and cerebrospinal fluid (CSF) amyloid-β42 level were significant difference among the SCD, MCI, and ADD groups. The frequencies of participants with amyloid pathology according to PET or CSF studies were 8.9%, 25.6%, 48.3%, and 90.0% in the CN, SCD, MCI, and ADD groups, respectively. According to ATN classification, A+/T+/N+ or A+/T+/N− was observed in 0%, 15.5%, 31.0%, and 78.3% in the CN, SCD, MCI, and ADD groups, respectively. The KBASE-V showed a clear difference according to the AD clinical spectrum in neuropsychological tests and AD biomarkers.

White matter hyperintensity (WMH) has been recognised as a surrogate marker of small vessel disease and is associated with cognitive impairment. We investigated the dynamic change in WMH in patients with severe WMH at baseline, and the effects of longitudinal change of WMH volume on cognitive decline and cortical thinning. Eighty-seven patients with subcortical vascular mild cognitive impairment were prospectively recruited from a single referral centre. All of the patients were followed up with annual neuropsychological tests and 3T brain magnetic resonance imaging. The WMH volume was quantified using an automated method and the cortical thickness was measured using surface-based methods. Participants were classified into WMH progression and WMH regression groups based on the delta WMH volume between the baseline and the last follow-up. To investigate the effects of longitudinal change in WMH volume on cognitive decline and cortical thinning, a linear mixed effects model was used. Seventy patients showed WMH progression and 17 showed WMH regression over a three-year period. The WMH progression group showed more rapid cortical thinning in widespread regions compared with the WMH regression group. However, the rate of cognitive decline in language, visuospatial function, memory and executive function, and general cognitive function was not different between the two groups. The results of this study indicated that WMH volume changes are dynamic and WMH progression is associated with more rapid cortical thinning.

Purpose: We assessed the effect of home-based cognitive intervention (HCI) on cognitive function along with brain metabolism by 18F-FDG PET in patients with amnestic MCI (aMCI). Patients and methods: Fifty-seven patients with aMCI from three hospitals were randomized (30 HCI, 27 control). For 12 weeks, subjects received HCI. Thirty-two subjects (15 HCI, 17 control) underwent brain 18-F-FDG-PET imaging at baseline and at 12 and 24 weeks. Results: The HCI group showed significant improvement in the scores of the Controlled Oral Word Association Test (COWAT) 12 and at 24 weeks. Significant brain metabolic changes by 18F-FDG PET were not observed. Conclusion: The current study suggests that HCI was effective in improving general cognition along with frontal executive function in patients with aMCI.