Unobtrusive Monitoring of Simulated Physical Weakness Using Fine-Grained Behavioral Features and Personalized Modeling

TL;DR

This work demonstrates that unobtrusive RGB-D camera monitoring of daily sitting and relaxing activities, combined with fine-grained behavioral features and personalized Bayesian Network modeling, can accurately detect simulated weakness and explain behavioral changes. By exploring multiple time scales and activity selections, the study finds that a $300\text{s}$ window often yields the best classification performance, achieving $0.97$ daily-level accuracy, with personalized feature sets outperforming universal ones. The approach provides interpretable insights into which movement and inactivity features, and which activities, are most indicative of weakness for individuals. These findings support scalable, privacy-preserving home monitoring for early detection of health deterioration in older adults, while highlighting the need for personalized models and larger-scale validation.

Abstract

Aging and chronic conditions affect older adults' daily lives, making early detection of developing health issues crucial. Weakness, common in many conditions, alters physical movements and daily activities subtly. However, detecting such changes can be challenging due to their subtle and gradual nature. To address this, we employ a non-intrusive camera sensor to monitor individuals' daily sitting and relaxing activities for signs of weakness. We simulate weakness in healthy subjects by having them perform physical exercise and observing the behavioral changes in their daily activities before and after workouts. The proposed system captures fine-grained features related to body motion, inactivity, and environmental context in real-time while prioritizing privacy. A Bayesian Network is used to model the relationships between features, activities, and health conditions. We aim to identify specific features and activities that indicate such changes and determine the most suitable time scale for observing the change. Results show 0.97 accuracy in distinguishing simulated weakness at the daily level. Fine-grained behavioral features, including non-dominant upper body motion speed and scale, and inactivity distribution, along with a 300-second window, are found most effective. However, individual-specific models are recommended as no universal set of optimal features and activities was identified across all participants.

Figures (11)

• Figure 1: A compact system designed for monitoring older adults in their homes employs an RGB-D camera and a computer processor. This system prioritizes privacy by discarding image/video data after extracting the necessary information. Motion, inactivity features, and environmental context are extracted in real time. Detected movement pixels are shown in red.
• Figure 2: Example of two features that differ among activities and between health states for a participant in all activities. The left two plots show how the feature values vary according to the activity (Act.). The right two plots show how the feature values vary (slightly) between the Normal and Weakness health states (Hea.).
• Figure 3: Structure of the Bayesian Network. Continuous variables are circles and discrete variables are rectangles. Shaded nodes represent observable features. The unobservable 'Goal' node is omitted. The network can be extended to a dynamic model by adding temporal links.
• Figure 4: F1-scores for inferring health states on all monitoring records with three classifiers (Bayesian Network (BN), Random Forest (RF), and Supportive Vector Machine (SVM)). allAct: classify with all activities; actUnk: classify with all activities with activity labels unknown; + TWin: classify with all activities with optimal temporal windows for feature extraction; + actSel: classify with optimally selected combination of activities (>40% data coverage, details are given in Section \ref{['sec:im']} Implementation Details.); + actSel + TWin: classify with optimally selected combination of activities and optimal temporal windows for feature extraction.
• Figure 5: F1-scores of classifying health states using the Bayesian Network across different temporal windows (30 to 1200 seconds) and aggregated for different time spans (record-level, 8-hour level, and daily-level) for five participants (P1--P5).