Journal for the measurement of physical behaviour最新文献

Background: Machine learning has been used for classification of physical behavior bouts from hip-worn accelerometers; however, this research has been limited due to the challenges of directly observing and coding human behavior "in the wild." Deep learning algorithms, such as convolutional neural networks (CNNs), may offer better representation of data than other machine learning algorithms without the need for engineered features and may be better suited to dealing with free-living data. The purpose of this study was to develop a modeling pipeline for evaluation of a CNN model on a free-living data set and compare CNN inputs and results with the commonly used machine learning random forest and logistic regression algorithms.

Method: Twenty-eight free-living women wore an ActiGraph GT3X+accelerometer on their right hip for 7 days. A concurrently worn thigh-mounted activPAL device captured ground truth activity labels. The authors evaluated logistic regression, random forest, and CNN models for classifying sitting, standing, and stepping bouts. The authors also assessed the benefit of performing feature engineering for this task.

Results: The CNN classifier performed best (average balanced accuracy for bout classification of sitting, standing, and stepping was 84%) compared with the other methods (56% for logistic regression and 76% for random forest), even without performing any feature engineering.

Conclusion: Using the recent advancements in deep neural networks, the authors showed that a CNN model can outperform other methods even without feature engineering. This has important implications for both the model's ability to deal with the complexity of free-living data and its potential transferability to new populations.

Little is known about how sedentary behaviour (SB) metrics derived from hip-worn and thigh-worn accelerometers agree for older adults. Thigh-worn activPAL micro monitors were concurrently worn with hip-worn ActiGraph GT3X+ accelerometers (with SB measured using the 100 count-per-minute (cpm) cut-point; ActiGraph_100cpm) by 953 older adults (age 77±6.6, 54% women) for 4-to-7 days. Device agreement for sedentary time and 5 SB pattern metrics was assessed using mean error and correlations. Logistic regression tested associations with 4 health outcomes using standardized (i.e., z-scores) and unstandardized SB metrics. Mean errors (activPAL-ActiGraph_100cpm) and 95% limits of agreement were: sedentary time -54.7(-223.4,113.9) min/d; time in 30+ minute bouts 77.6(-74.8,230.1) min/d; mean bout duration 5.9(0.5,11.4) min; usual bout duration 15.2(0.4,30) min; breaks in sedentary time -35.4(-63.1,-7.6) breaks/d; and alpha -0.5(-0.6,-0.4). Respective Pearson correlations were: 0.66, 0.78, 0.73, 0.79, 0.51, 0.40. Concordance correlations were: 0.57, 0.67, 0.40, 0.50, 0.14, 0.02. The statistical significance and direction of associations was identical for ActiGraph_100cpm and activPAL metrics in 46 of 48 tests, though significant differences in the magnitude of odds ratios were observed among 9 of 24 tests for unstandardized and 2 of 24 for standardized SB metrics. Caution is needed when interpreting SB metrics and associations with health from ActiGraph_100cpm due to the tendency for it to overestimate breaks in sedentary time relative to activPAL. However, high correlations between activPAL and ActiGraph_100cpm measures and similar standardized associations with health outcomes suggest that studies using ActiGraph_100cpm are useful, though not ideal, for studying SB in older adults.