Real-time estimation of daily physical activity intensity by a triaxial accelerometer and a gravity-removal...

Area of Science:

• Biomedical Engineering
• Kinesiology
• Physical Activity Measurement

Background:

• Accurate assessment of physical activity intensity is crucial for health monitoring and research.
• Existing methods for classifying physical activity intensity using accelerometers have limitations.
• The gravity-removal physical activity classification algorithm (GRPACA) offers a novel approach.

Purpose of the Study:

• To develop and validate a new model for immediate estimation of daily physical activity intensities.
• To utilize a triaxial accelerometer and the GRPACA for enhanced activity classification.
• To compare the accuracy of the GRPACA model with existing classification methods.

Main Methods:

• Sixty-six subjects were divided into validation (n=44) and cross-validation (n=22) groups.
• Subjects performed fourteen activities while wearing a triaxial accelerometer.
• Energy expenditure was measured using indirect calorimetry, with intensity expressed as metabolic equivalents (MET).

Main Results:

• Strong correlations were found between measured MET and filtered synthetic accelerations for household (r=0.907) and locomotive (r=0.961) activities in the validation group.
• GRPACA-based linear regression models accurately estimated MET for both activity types in the cross-validation group.
• Sedentary activities were also accurately estimated using a specific linear regression model.

Conclusions:

• The GRPACA, combined with a triaxial accelerometer, enables more accurate and immediate estimation of daily physical activity intensities.
• This method surpasses the accuracy of previously reported cut-off classification models.
• The GRPACA approach is suitable for field investigations and personal self-monitoring.