Key result
Raw-based Hees algorithm outperforms epoch-based methods for detecting accelerometer non-wear time, reaching ~0.78 F1 score.
Why the study?
Accurate detection of accelerometer non-wear time is crucial for calculating physical activity summary statistics, but the performance of raw and epoch-based non-wear algorithms needed evaluation.
Observational (n=583)
With tuned hyperparameters, epoch-based non-wear algorithms can achieve classification performance comparable to raw-based algorithms for detecting accelerometer wear time.
Answers are grounded in this paper and cite their sources.
May refine accelerometer-based activity monitoring in research; leaves open whether tuned epoch-based methods achieve comparable performance.
Accurate detection of accelerometer non-wear time is crucial for calculating physical activity summary statistics. In this study, we evaluated three epoch-based non-wear algorithms (Hecht, Troiano, and Choi) and one raw-based algorithm (Hees). In addition, we performed a sensitivity analysis to provide insight into the relationship between the algorithms' hyperparameters and classification performance, as well as to generate tuned hyperparameter values to better detect episodes of wear and non-wear time. We used machine learning to construct a gold-standard dataset by combining two accelerometers and electrocardiogram recordings. The Hecht and Troiano algorithms achieved poor classification performance, while Choi exhibited moderate performance. Meanwhile, Hees outperformed all epoch-based algorithms. The sensitivity analysis and hyperparameter tuning revealed that all algorithms were able to achieve increased classification performance by employing larger intervals and windows, while more stringently defining artificial movement. These classification gains were associated with the ability to lower the false positives (type I error) and do not necessarily indicate a more accurate detection of the total non-wear time. Moreover, our results indicate that with tuned hyperparameters, epoch-based non-wear algorithms are able to perform just as well as raw-based non-wear algorithms with respect to their ability to correctly detect true wear and non-wear episodes.
No takes yet. Share an insight, caveat, or question.
Syed et al. (2020) conducted an observational in Healthy adults (general population) (n=583). Raw and epoch non-wear algorithms (Hecht, Troiano, Choi, Hees) vs. Gold standard dataset (machine learning combined with two accelerometers and ECG) was evaluated on Classification performance (F1 score) for detecting non-wear time. The raw-based Hees algorithm outperformed epoch-based algorithms in detecting true non-wear time, achieving an F1 score of 0.78 on the total data.
Synapse has enriched 10 related papers. Consider them for comparative context: