Posture and Movement Classification: The Comparison of Tri-Axial Accelerometer Numbers and Anatomical Placement

Abstract

Patient compliance is important when assessing movement, particularly in a free-living environment when patients are asked to don their own accelerometers. Reducing the number of accelerometers could increase patient compliance. The aims of this study were (1) to determine and compare the validity of different accelerometer combinations and placements for a previously developed posture and dynamic movement identification algorithm. Custom-built activity monitors, each containing one tri-axial accelerometer, were placed on the ankles, right thigh, and waist of 12 healthy adults. Subjects performed a protocol in the laboratory including static orientations of standing, sitting, and lying down, and dynamic movements of walking, jogging, transitions between postures, and fidgeting to simulate free-living activity. When only one accelerometer was used, the thigh was found to be the optimal placement to identify both movement and static postures, with a misclassification error of 10%, and demonstrated the greatest accuracy for walking/fidgeting and jogging classification with sensitivities and positive predictive value (PPVs) greater than 93%. When two accelerometers were used, the waist-thigh accelerometers identified movement and static postures with greater accuracy than the thigh-ankle accelerometers (with a misclassification error of 11% compared to 17%). However, the thigh-ankle accelerometers demonstrated the greatest accuracy for walking/ fidgeting and jogging classification with sensitivities and PPVs greater than 93%. Movement can be accurately classified in healthy adults using tri-axial accelerometers placed on one or two of the following sites: waist, thigh, or ankle. Posture and transitions require an accelerometer placed on the waist and an accelerometer placed on the thigh.

Step detection using multi- versus single tri-axial accelerometer-based systems

“…Two outliers occurred with the waist and ankle combination (figure 4e). While a single thigh AM produced the least accurate results for step detection (figure 4b), it yielded the most accurate results for detecting time spent walking and jogging as found in a parallel study by the authors using the same data recordings (Fortune et al , 2014b). The mean overestimation of step counts of less than 3% using the single waist AM (figure 4a) was due to false positives steps at the ends of segments as the duration of upright movement detection was overestimated (Fortune et al , 2014b).…”

“…The SMA thresholds for detecting ‘walking/fidgeting’ or ‘jogging’ activity when using the thigh or ankle acceleration data were calculated as ratios of the waist acceleration data and are presented in (Fortune et al , 2014b). When using the single waist or single ankle AMs, angle estimations were used to identify lying down postures from upright postures.…”

“…The algorithm’s validity to detect steps was evaluated by comparison to video recordings. The algorithm used in this study is built on previous work and data by our research group on postural and activity detection (Lugade et al , 2014; Fortune et al , 2014b), and step count (Fortune et al , 2014a) and is additionally validated for an older adult population.…”

Sedentary behavior compensation to 1‐year exercise RCT in patients with type 2 diabetes

“…We observed no differences in MVPA between groups likely due to the inability of the accelerometers to detect the increased physical activity energy expenditure associated with stationary bicycling and resistance training in the machines. Underestimation of movement and physical activity energy expenditure by waist mounted accelerometry during stationary activities is well documented . Despite this limitation, it is unlikely this had a major influence on our primary aim examining a potential intervention effect on increasing sedentary behavior indicators.…”

Predicting daily gait behaviors after anterior cruciate ligament surgery: A case study

“…SMA has been used to quantify the amount of gait activity [15]. It is similar to other measures used to quantify human activity [18], [19].…”