Risk assessment for work-related musculoskeletal injury using wearable sensors

• Author / Creator

  Beltran Martinez, Karla

• Work-related musculoskeletal disorders are a worldwide problem that affects millions of people every year, and they have a cost of billions of dollars that keeps increasing over the years. One of the main causes of musculoskeletal disorders is fatigue. One of the most used and relievable methods to detect fatigue is with electromyography (EMG) signal, and even though the results obtained with this method are good enough for some applications, the use of these sensors can be time-consuming, expensive, and inconvenient. Currently, there is no objective and accurate measure of fatigue for long-term measurement in real-world working scenarios.
  This thesis investigated the use of wearable inertial measurement units (IMU) for body posture, fatigue, and ergonomic risk assessments in long-term and real-world working conditions. To this end, we proposed to measure and detect fatigue in-field using only kinematic information with an established parameter, i.e., Rapid Entire Body Assessment (REBA), and a novel kinematic parameter, i.e., K-score introduced in this study. To investigate the validity of these two parameters for muscle fatigue detection, we performed an experimental study involving a material handling task with 10 able-bodied participants. The fatigue was measured with three methods: a Borg Rating of Perceived Exertion scale RPE, EMG signal, and the two kinematic parameters measured with IMUs. The results showed that REBA did not have a significant correlation (p>0.05) with the EMG signal amplitude affected by muscle fatigue. The lack of correlation between REBA and EMG amplitude could be related to the limited resolution of the REBA score. Our introduced kinematic parameter (K-score) is a function of body joint angles but has a higher resolution than REBA and could tackle this challenge. K-score showed a correlation coefficient of ρ = 0.21 (p <0.05) with EMG amplitude, which validated its use for fatigue detection in repetitive tasks.
  Furthermore, we investigated the difference of muscle fatigue for three work-rest schedules: (1) a typical 30-min trial without any breaks, (2) adding two one-minute micro-breaks in between the trial, and (3) adding two breaks in between to perform specific stretching exercises. EMG signal and K-score were recorded for muscle fatigue characterization. Both parameters showed a significant difference among the trials using a multiple comparison test. We concluded that 1) micro-breaks can have a meaningful muscle fatigue reduction in working conditions, which may contribute to a reduced risk of WMSD, and 2) K-score has the potential to detect and characterize muscle fatigue, and their measurements using an IMU could be a substitute for EMG measurement that is challenging in the long-term.
  Lastly, we present a validation for IMUs to properly measure ergonomic risk assessments compared to camera-based motion capture for five minutes and in a material handling task. We measured the accuracy of IMUs to measure 1) 3D joint angles, and 2) two kinematic parameters for ergonomic risk assessment: REBA and K-score. IMUs were able to measure the 3D joint angles of the shoulder, trunk, elbow, and knee with an RMSE of less than 4° and the neck joint angles with a mean RMSE of 6.2°. Furthermore, REBA and K-score based on joint angles measured with IMUs and motion capture were compared using Cohen’s Kappa coefficient, and demonstrated to have a “substantial agreement”.
  In summary, these studies demonstrated the accuracy of wearable IMUs for the measurement of body joint angles, ergonomic scores, and muscle fatigue in long-term and real-world conditions.

• Subjects / Keywords

  • Wearable sensors
  • Ergonomics
  • Muscle Fatigue

• Graduation date

  Spring 2021

• Type of Item

  Thesis

• Degree

  Master of Science

• DOI

  https://doi.org/10.7939/r3-62cq-n921

• License

  This thesis is made available by the University of Alberta Libraries with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.