Classifying physical activity levels using Mean Amplitude Deviation in adults using a chest worn accelerometer:...

Area of Science:

• Wearable sensor technology
• Biomedical engineering
• Physical activity monitoring

Background:

• Wearable accelerometers offer objective remote monitoring of physical activity (PA), surpassing self-reporting.
• Reliably detecting ambulatory activity is crucial for accelerometer data validity.
• Mean Amplitude Deviation (MAD) is a strong metric for analyzing accelerometer data and defining activity cut-points.

Purpose of the Study:

• To calibrate the accelerometer function of the VivaLink ECG patch.
• To determine the Mean Amplitude Deviation (MAD) cut-point for differentiating sedentary and ambulatory activities using chest-worn accelerometers.

Main Methods:

• Healthy volunteers performed 9 simulated free-living activities while wearing a VivaLink ECG Patch.
• Generalized Linear Mixed Models were used to determine activity cut-points based on MAD values.
• Receiver Operating Characteristic (ROC) curves analyzed the sensitivity and specificity of the MAD cut-point.

Main Results:

• The optimal MAD cut-point for distinguishing sedentary from ambulatory activity was 47.73mG.
• ROC curve analysis showed an Area Under the Curve of 0.99 (p < 0.001).
• The determined cut-point achieved 98% sensitivity and 100% specificity.

Conclusions:

• The identified MAD cut-point effectively categorizes sedentary and ambulatory activities in healthy adults.
• This method offers a low-burden approach for community-based PA monitoring.
• The findings facilitate future research integrating PA monitoring with other physiological data from chest-worn sensors.