Comparison of Different Signal Processing Methodologies and Their Impact on the Range of Acceleration Amplitudes Experienced by Preschool-Aged Children

ABSTRACT

This study explored 1) the effect of applying an autocalibration algorithm on accelerometer data analyzed using different signal processing techniques and 2) how these techniques changed the signal composition. Using a dataset of preschoolers (n = 137; 4.31 ± 0.87 years; Actigraph accelerometers (100 Hz) on right hip 24 hours/7 days) the vector magnitude (VM) was calculated as a reference; highpass, bandpass and lowpass filters were applied and the “Euclidean Norm Minus One” (ENMO) was calculated. Amplitude probability distribution functions were created to compare the 1) uncalibrated versus calibrated data and, 2) amplitude distribution across the techniques. Results showed that the calibration correction significantly impacted ENMO, VM lowpass and VM data, but the impact on VM highpass and bandpass data was minimal. Comparing across the techniques, results showed the lowest outputs were derived according to ENMO, followed by the VM highpass/VM bandpass and the VM lowpass/VM which yielded very similar results.

KEYWORDS:

• Assessment
• evaluation
• measurement

Acknowledgments

The authors would like to thank Dr. Stephen Brown for his suggestions on on an earlier version of this manuscript.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This project was funded by the Canada Foundation for Innovation (John R. Evans Leaders Fund awarded to DWL Ma; Grant #36619) and the Canadian Institutes of Health Research (Project grant #376067 awarded to J. Haines), as well as by an Ontario Graduate Scholarship (awarded to H. Coyle-Asbil).