A method to estimate free-living active and sedentary behavior from an accelerometer

Abstract

Introduction: Methods to estimate physical activity (PA) and sedentary behavior (SB) from wearable monitors need to be validated in free-living settings.

Purpose: The purpose of this study was to develop and validate two novel machine-learning methods (Sojourn-1 Axis [soj-1x] and Sojourn-3 Axis [soj-3x]) in a free-living setting.

Methods: Participants were directly observed in their natural environment for 10 consecutive hours on three separate occasions. Physical activity and SB estimated from soj-1x, soj-3x, and a neural network previously calibrated in the laboratory (lab-nnet) were compared with direct observation.

Results: Compared with lab-nnet, soj-1x and soj-3x improved estimates of MET-hours (lab-nnet: % bias [95% confidence interval] = 33.1 [25.9 to 40.4], root-mean-square error [RMSE] = 5.4 [4.6-6.2]; soj-1x: % bias = 1.9 [-2.0 to 5.9], RMSE = 1.0 [0.6 to 1.3]; soj-3x: % bias = 3.4 [0.0 to 6.7], RMSE = 1.0 [0.6 to 1.5]) and minutes in different intensity categories {lab-nnet: % bias = -8.2 (sedentary), -8.2 (light), and 72.8 (moderate-to-vigorous PA [MVPA]); soj-1x: % bias = 8.8 (sedentary), -18.5 (light), and -1.0 (MVPA); soj-3x: % bias = 0.5 (sedentary), -0.8 (light), and -1.0 (MVPA)}. Soj-1x and soj-3x also produced accurate estimates of guideline minutes and breaks from sedentary time.

Conclusions: Compared with the lab-nnet algorithm, soj-1x and soj-3x improved the accuracy and precision in estimating free-living MET-hours, sedentary time, and time spent in light-intensity activity and MVPA. In addition, soj-3x is superior to soj-1x in differentiating SB from light-intensity activity.

Figure 1. Measuring free-living physical activity and sedentary behavior

Bottom and middle panels show 2-min 30-sec of second-by-second counts from the vertical acceleration signal. Top panel shows observer-identified activities. Using the lab-nnet and simple regression approaches the five distinct activities are grouped into minute intervals (bottom panel), resulting in inaccurate MET estimates. In free-living environments it may be more appropriate to identify where bouts of activity start and stop (middle panel) and estimate METs for specific activity bouts.

Figure 2

Lab-Nnet, Soj-1x and Soj-3x estimates for each participant. Model estimates for each participant compared to direct observation. The closer the point falls to the line of identity, the closer the estimate is to direct observation. Lab-nnet: open squares, soj-1x: open triangles, soj-3x: filled circles.

Figure 3

Freedson 1998, Crouter 2006 and Crouter 2010 estimates for each participant. Model estimates for each participant compared to direct observation. The closer the point falls to the line of identity, the closer the estimate is to direct observation. Freedson 1998: open squares, Crouter 2006: open triangles, Crouter 2010: filled circles.

Figure 4

Second-by-second counts from vertical, anterior-posterior and medial-later axes (top). Corresponding Soj-1x and Soj-3x estimates compared to direct observation (bottom). These data illustrate an example of when the additional information from the anterior-posterior and medial-lateral axes help soj-3x correctly identify light intensity activity where soj-1x inaccurately estimates this activity as sedentary using information from the vertical axes alone.