Perceived fatigue progression tracking during manual handling tasks using sEMG recordings.

The study included only eight male participants, limiting generalizability to women or diverse populations. Experiments were conducted in a controlled laboratory setting, which may not reflect the variability and distractions of real-world workplaces. The deep learning model's 69% accuracy, while promising, is not yet sufficient for reliable clinical or occupational decision-making without further validation. The study relied on self-reported fatigue (RPE) as the ground truth, which is subjective and can be influenced by factors other than physical muscle fatigue.

The study was funded by the Natural Sciences and Engineering Research Council of Canada and Alberta Innovates. No specific conflicts of interest regarding commercial entities were declared in the provided text.

Do not interpret this as evidence that wearable sEMG devices are ready for clinical use or that they can accurately diagnose fatigue in individual patients. The study was a small pilot with a homogeneous male sample, and the 69% classification accuracy is not high enough for reliable individual monitoring in real-world settings.