Related Experiment Video
Updated: Sep 15, 2025

Methodology for Establishing a Community-Wide Life Laboratory for Capturing Unobtrusive and Continuous Remote Activity and Health Data
Published on: July 27, 2018
Quantifying Mild Cognitive Impairments in Older Adults Using Multi-modal Wearable Sensor Data in a Kitchen
Bonwoo Koo1, Ibrahim Bilau2, Amy D Rodriguez3
1KAIST, Daejeon, South Korea.
Wearable sensors can detect Mild Cognitive Impairment (MCI) by analyzing kitchen activities. This technology shows promise for early screening of cognitive decline in older adults.
Area of Science:
- Neuroscience
- Gerontology
- Human-Computer Interaction
Background:
- Mild Cognitive Impairment (MCI) screening often relies on traditional methods, with limited research on behavioral markers in daily living settings.
- Existing wearable studies for MCI primarily focus on gait analysis in controlled environments.
- Kitchen activities, crucial for daily living, are known to be affected by visuospatial deficits common in MCI.
Purpose of the Study:
- To investigate the use of wrist and eye-tracking wearable sensors for quantifying kitchen activities in individuals with MCI.
- To explore the potential of multimodal sensing for differentiating individuals with MCI from those with normal cognition.
- To identify behavioral markers associated with cognitive decline during a common daily task.
Main Methods:
- Collected multimodal data using wrist and eye-tracking sensors from 19 older adults (11 with MCI, 8 with normal cognition) performing a yogurt bowl preparation task.
- Developed a multimodal analysis model to classify participants based on their sensor data.
- Performed feature importance analysis to understand which behavioral markers are most associated with MCI.
Main Results:
- The multimodal analysis model achieved a 74% F1 score in classifying individuals with MCI from those with normal cognition.
- Feature importance analysis revealed associations between weaker upper limb motor function and delayed eye movements with cognitive decline.
- Findings align with previous research indicating motor and visual-behavioral changes in MCI.
Conclusions:
- This pilot study demonstrates the feasibility of using wearable sensors to monitor behavioral markers of MCI during everyday kitchen activities.
- The findings suggest that multimodal sensing can aid in the early detection of cognitive decline in real-world settings.
- Further large-scale validation studies in home environments are warranted to confirm these promising results.
More Related Videos
04:13Using a Real-Time Locating System to Measure Walking Activity Associated with Wandering Behaviors Among Institutionalized Older Adults
Published on: February 8, 2019
08:45Measuring the Kinematics of Daily Living Movements with Motion Capture Systems in Virtual Reality
Published on: April 5, 2018