Application Of The ActiGraph GT9X IMU For The Assessment Of Turning During Walking And Running

Abstract

Rotational movements, such as turning, can significantly increase energy expenditure (EE) during ambulatory activity. Gyroscope and magnetometer sensors can quantify rotational motion, which provides additional information on movement beyond linear acceleration that is provided by only using an accelerometer. PURPOSE: The purpose of this study was to examine the use of the ActiGraph GT9X gyroscope and magnetometer for detecting turns and quantifying turn degree during walking and running. METHODS: Participants (N=17) completed pivot trials, treadmill walking and running (TM; 3 to 6 mph) and four turn conditions (i.e. 45°, 90°, 135°, and 180°) during over- ground walking and running (OG). Pivot and TM trials were completed for 1-min and 6-min, respectively. Turn frequency was constant (10 turns/min) for all OG trials. A GT9X was placed on the left hip and a Cosmed K4b2 was used to measure EE. Raw GT9X gyroscope and magnetometer data were processed through various low-pass filter frequencies (0.25 Hz to 2.0 Hz). TM and pivot trials were used to develop thresholds for turn detection using the gyroscope and magnetometer data and the OG trials were used for cross-validation. K4b2 data (VO2) were averaged over 30-s periods and converted to relative VO2 (ml/kg.min). Linear mixed models were used to compare actual and predicted number of turns, measured and predicted turn degree, and differences in VO2 across OG conditions. RESULTS: There were no main effects for speed or turn condition on turn detection when filtering the gyroscope at 0.25 Hz (p>0.05). A speed main effect was present when filtering the magnetometer at 0.75 Hz (p<0.001). 0.25 Hz (gyroscope) and 0.75 Hz (magnetometer) filters resulted in 100.5%±4.4% and 96.9%±45.4% of turns detected, respectively. Using the gyroscope, turn degree was estimated to within approximately 2.2° for all turn conditions (p<0.001). In general, the VO2 of walking and running was significantly greater during 135° and 180° turn conditions compared to 0-90° turn conditions (p<0.05). CONCLUSION: The GT9X gyroscope, when low-pass filtered at 0.25 Hz, can be used to detect the number of turns and estimate turn degree. The magnetometer was only useful for detecting the number of turns. Future work should explore the gyroscope use for turn detection during activities other than walking and running.