ActivityAware: An App for Real-Time Daily Activity Level Monitoring on the Amulet Wrist-Worn Device

Area of Science:

• Biomedical Engineering
• Health Informatics
• Wearable Technology

Background:

• Physical activity is crucial for reducing risks of cardiovascular disease, hypertension, and obesity.
• Monitoring daily activity levels aids in self-management and targeted health interventions.
• Older adults with obesity particularly benefit from regular physical activity to prevent long-term disability.

Purpose of the Study:

• To present ActivityAware, a novel application for the Amulet wrist-worn device.
• To enable continuous, real-time measurement of individual daily activity levels (sedentary, moderate, vigorous).
• To develop and validate an activity-level detection model for improved health monitoring.

Main Methods:

• Developed an activity-level detection model using a Support Vector Machine (SVM) classifier.
• Collected continuous acceleration data from the Amulet wrist device.
• Trained and validated the SVM model using data from a user study involving sit, stand, lay, walk, and run activities.

Main Results:

• Achieved preliminary accuracies up to 98% using 10-fold and leave-one-subject-out cross-validation with 14 subjects.
• The ActivityAware app demonstrated a projected battery life of up to 4 weeks.
• The system successfully classifies activity levels and logs data for analysis.

Conclusions:

• The ActivityAware app shows promise for accurate real-time activity-level monitoring.
• This technology can inform the development of personalized interventions to improve individual health.
• Wearable technology offers a viable solution for continuous health behavior tracking.