Related Experiment Video
Updated: Sep 18, 2025
11:21
Methodology for Establishing a Community-Wide Life Laboratory for Capturing Unobtrusive and Continuous Remote Activity and Health Data
Published on: July 27, 2018
8.3K
Exploring Factors Related to Social Isolation Among Older Adults in the Predementia Stage Using Ecological Momentary
Bada Kang1,2, Min Kyung Park1, Jennifer Ivy Kim1
1Mo-Im Kim Nursing Research Institute, Yonsei University College of Nursing, Seoul, Republic of Korea.
Journal of Medical Internet Research
|June 23, 2025
Summary
Machine learning models identified physical movement as key to social interaction and sleep quality to loneliness in older adults at risk of dementia. These findings aid in preventing cognitive decline through early detection of social isolation factors.
Area of Science:
- Gerontology
- Artificial Intelligence
- Cognitive Science
Background:
- Dementia poses a growing economic burden globally.
- Social isolation and loneliness are significant risk factors for dementia and negatively impact cognitive function.
- Identifying factors in predementia stages (subjective cognitive decline, mild cognitive impairment) is crucial for intervention.
Purpose of the Study:
- Develop and validate machine learning models to identify factors related to social interaction frequency.
- Explore factors associated with loneliness levels in older adults in predementia stages.
Main Methods:
- Utilized mobile ecological momentary assessment and wearable actigraphy data from 99 community-dwelling older adults (65+ years) in predementia stages.
- Assessed social interaction frequency and loneliness 4 times daily over 2 weeks.
- Employed logistic regression, random forest, Gradient Boosting Machine, and Extreme Gradient Boosting models.
Main Results:
- Random forest model best identified factors for low social interaction frequency (accuracy 0.849).
- Gradient Boosting Machine model best identified factors for high loneliness levels (accuracy 0.838).
- Physical movement was linked to social interaction; sleep quality was linked to loneliness.
Conclusions:
- Machine learning models effectively detect social isolation risks in older adults with predementia.
- Physical movement and sleep quality emerge as critical factors influencing social interaction and loneliness.
- This approach can help prevent cognitive and physical decline in high-risk older adults.
