EMG-BASED DETECTION OF PERFORMANCE FATIGUE IN MULTI-ACTIVITY MANUAL HANDLING TASKS
Authors
Armin Bonakdar, Karla Beltran Martinez, Ali Golabchi, Mahdi Tavakol, Hossein Rouhani
Abstract
Work-related musculoskeletal disorders (WMSDs) represent a significant global concern, affecting millions annually1 . One of the primary contributors to WMSDs is performance fatigue2 , and understanding its temporal progression during manual handling tasks is crucial for designing effective interventions to prevent WMSDs. We hypothesize that this can be done through monitoring muscle activities using electromyography (EMG) sensors3 . While previous studies have not extensively explored fatigue detection during multi-activity labour tasks, this study aims to investigate performance fatigue during prolonged, multi- activity manual handling tasks via EMG analysis.
Methods
Five able-bodied male participants (age: 24 ± 2 years, body mass: 73 ± 11 kg, body height: 179 ± 4 cm) engaged in an experiment simulating a material handling task. This task involved repetitive actions: lifting a 7.2 kg box from a table 15 cm in height (lifting task), transferring it to a table 75 cm in height (carrying task), and then lowering it back to the initial table (lowering task). Participants were instructed to continue the tasks until they reached the Borg scale fatigue level of 9 out of 10. EMG sensors were placed on the right side due to the symmetrical nature of the task. The Biceps Femoris, Lateral Gastrocnemius, Erector Spinae Iliocostalis, and Erector Spinae Longissimus were selected based on their substantial involvement in this specific activity. EMG readouts were filtered using a 4th-order band-pass Butterworth filter (10–500 Hz), and each muscle's maximum voluntary contraction was obtained at the start of the trial. Muscle activations were assessed by measuring the root mean square (RMS) value and median frequency during each repetition to observe the effect of fatigue. As repetitions progress, it is hypothesized that the RMS of EMG signals increases due to increased recruitment of motor units4, while a decrease is anticipated in the median frequency of the EMG signals5.
Results and Conclusion
The positive correlations between RMS values of EMG and fatigue levels indicate an increase in muscle activation as fatigue sets in due to the need for the recruitment of additional motor units to sustain task performance. Notably, this increase was most evident during the lifting and lowering phases, with an average Spearman correlation of 0.29 and 0.26, compared to the carrying phase, with an average of 0.18 (Figure 1).
References
[1] Lopez Del Puerto et al., 2014, 50th ASC Annual International Conference Proceedings.
[2] Ricci et al., 2007, Journal of Occupational an Environmental Medicine., 1–10
[3] Cifrek et al., 2009 Clin. Biomech. 327–340
[4] Makaram et al., 2021, IEEE Transactions on Instrumentation and Measurement, 1-8.
[5] van Sluijs et al., 2023, Wearable Technologies.