Effect of epoch length on intensity classification and on accuracy of measurement under controlled conditions on treadmill: Towards a better understanding of accelerometer measurement

Abstract

The aim of this study was to analyze the effect of epoch length on intensity classification during continuous and intermittent activities. Methods Ten active students exercised under controlled conditions on a treadmill for four 5-min bouts by combining two effort intensities (running and walking) and two physical activity (PA) patterns (continuous or intermittent). The testing session was designed to generate a known level of moderate to vigorous PA (MVPA) for each condition. These PA levels were used as criterion measures to compare with the accelerometer measures. Data obtained from the accelerometer were reintegrated into 1-sec, 10-sec, 30-sec and 60-sec epochs. Equivalence testing was used to examine measurement agreements between MVPA values obtained with the different epochs and the reference values. Mean absolute percent errors (MAPE) were also calculated to provide an indicator of overall measurement error. Results During the intermittent conditions, only the value obtained with the 1-sec epoch was significantly equivalent to the reference value. With longer epochs the difference increased for both intermittent conditions but in an opposite way: with longer epochs, MVPA decreased during walking but increased during running. Regarding the measurement accuracy, the pattern of variations according to the epoch length selected during the intermittent conditions was identical between walking and running: MAPE increased with the increase in epoch length. MAPE remained low only for the 1-sec epoch (7.6% and 2.7% for walking and running, respectively), increased at 31.3% and 34% for the 10-sec epoch and until near 100% with the 30-and 60-sec epoch lengths.

Implementation of a deep learning model for automated classification of Aedes aegypti (Linnaeus) and Aedes albopictus (Skuse) in real time

“…Our results showed that both the number of epochs and the learning rate (LR) exhibited significant effects on the testing accuracy (p < 0.05), and when the number of epochs was 30 and the LR was 0.001, the model achieved an accuracy of 98.06 ± 1.02%. The number of epochs was the number of times that the entire training dataset was used for the algorithm learning process, and each epoch consisted of one or more batches that were used to tune the internal model parameter 26 . Theoretically, more epochs should result in higher accuracy 27 , although a longer runtime was required.…”

“…This was supported by the study of Ladds et al 34 , which investigated the classification of animal behavior using a super machine learning model and showed that fewer epochs performed better than more epochs in classifying animal behavior. Furthermore, Fabre et al’s 35 study tested more epochs, but this decreased the moderately vigorous physical activity and increased the percentage error of the accelerometer measurement. More epoch decreasing the performance of a model could be explained by the possibility of overfitting the model.…”

Key facets to build up eHealth and mHealth interventions to enhance physical activity, sedentary behavior and nutrition in healthy subjects – an umbrella review

“…Mainly including device-measured outcomes will lead to a more comprehensive picture of intervention effectiveness even though other challenges arise from that approach (e.g. comparison of different epoch lengths [ 69 ]). The most promising aspect of device-measured outcomes and accelerometry in particular is the assessment of valid PA and SB data in real-time, resulting in a variety of outcome parameters which have the potential to be easily compared throughout different studies [ 60 ].…”

Children’s Perceived and Actual Physical Activity Levels within the Elementary School Setting

“…The use of a 60 s epoch recording was chosen as this is the recommended way to assess the physical activity levels of children [ 33 , 34 , 35 ]; particularly in longitudinal data collection. However, caution is aired [ 55 ] in the use of a 60 s epoch for future research, due to the misclassification of intensity levels, particularly at vigorous levels, and the potential underestimation of time spent engaging in vigorous physical activity.…”