Related Experiment Video
Updated: Jan 9, 2026

A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
Published on: January 11, 2020
Validation of the Japanese version of MemScreen: a rapid screening tool for mild cognitive impairment
Ai Ikeda1,2, Hadrien Charvat1,2, Takeshi Tanigawa1
1Department of Public Health, Graduate School of Medicine, Juntendo University.
Abstract:
This study was to examine the validity of the Japanese version of MemScreen (MemScreen-J), a touchscreen MCI screening test. 20 patients with MCI aged 65-90 years at the Juntendo Tokyo Koto Geriatric Medical Center were recruited as cases in December 2023. Non-cases were recruited from local residents in Toon City, Ehime Prefecture in February 2024 and 40 residents, without a medical history of MCI, aged 58-84 years were included in the present study. MemScreen-J test, a self-administered screening test in the form of a digital application, downloadable on a tablet, was administered to participants to assess their cognitive function. Defining the group at high risk of MCI based on a MemScreen-J test score of 28 or lower achieved the best Youden index in the study sample, with a sensitivity of 0.75 and a specificity of 0.98. MemScreen-J appeared to be a valid screening tool among persons at the prodromal stage of dementia, given reasonably high accuracy in detection of MCI. This innovative neuropsychological test could be the first step in a diagnostic approach to cognitive complaints in a community, identifying persons at the preclinical stage of dementia.

