Comparison of the performance of the activPAL™ Professional physical activity logger to a discrete accelerometer-based activity monitor

Alan Godfrey, K. M. Culhane, Gerard Lyons

Research output: Contribution to journalArticlepeer-review

134 Citations (Scopus)

Abstract

The aim of this study was to assess the accuracy of the 'activPAL™ Professional' physical activity logger by comparing its output to that of a proven discrete accelerometer-based activity monitor during extended measurements on healthy subjects while performing activities of daily living (ADL). Ten healthy adults, with unrestricted mobility, wore both the activPAL™ and the discrete dual accelerometer (Analog Devices ADXL202)-based activity monitor that recorded in synchronization with each other. The accelerometer derived data were then compared to that generated by the activPAL™ and a complete statistical and error analysis was performed using a Matlab® program. This program determined trunk and thigh inclination angles to distinguish between sitting/lying, standing and stepping for the discrete accelerometer device and amount of time spent on each activity. Analysis was performed on a second-by-second basis and then categorized at 15 s intervals in direct comparison with the activPAL™ generated data. Of the total time monitored (approximately 60 h) the detection accuracies for static and dynamic activities were approximately 98%. In a population of healthy adults, the data obtained from the activPAL™ Professional physical activity logger for both static and dynamic activities showed a close match to a proven discrete accelerometer data with an offset of approximately 2% between the two systems.

Original languageEnglish
Pages (from-to)930-934
Number of pages5
JournalMedical Engineering and Physics
Volume29
Issue number8
Early online date28 Nov 2006
DOIs
Publication statusPublished - 1 Oct 2007

Fingerprint

Dive into the research topics of 'Comparison of the performance of the activPAL™ Professional physical activity logger to a discrete accelerometer-based activity monitor'. Together they form a unique fingerprint.

Cite this