In the paper “Examination of Social Participation in Older Adults Undergoing Frailty Health Checkups Using Deep Learning Models” by Professor Yokokawa’s group at Shinshu University, analyses were conducted using Multi-Sigma®.
The paper reports results analyzing how frailty in older adults restricts social participation by employing AI (machine learning) models.
In this analysis, Multi-Sigma® was used to build and analyze a deep neural network (DNN) to predict social participation among older adults who underwent frailty health checkups.
Approximately 300 older adults participated in the study, and three types of machine learning models were compared using data on 18 variables, including physical, cognitive, and social factors.
The DNN powered by Multi-Sigma® demonstrated well-balanced predictive performance and showed high sensitivity in identifying older adults who participate in social activities.
The findings published in the paper are expected to open new possibilities for initiatives aimed at preventing social isolation among older adults and extending healthy life expectancy.
[Paper Information]
Title: Examination of Social Participation in Older Adults Undergoing Frailty Health Checkups Using Deep Learning Models
Authors: Yoshiharu Yokokawa et al.
Journal: Geriatrics 2025, 10(5), 124
Year: 2025
DOI: https://doi.org/10.3390/geriatrics10050124
