Statistical Prediction of Hand Force Exertion Levels in a Simulated Push Task using Posture Kinematics

Area of Science:

• Biomechanics
• Ergonomics
• Wearable Technology

Background:

• Estimating force exertion in occupational settings is crucial for assessing biomechanical risk factors.
• Traditional methods for measuring force can be intrusive and limit natural movement.
• Body posture kinematics offer a potential non-invasive approach to infer force levels.

Purpose of the Study:

• To investigate the use of body posture kinematics from wearable inertial sensors for estimating force exertion levels.
• To develop a predictive model correlating body posture with force magnitude in isometric pushing and pulling tasks.
• To assess the feasibility of using wearable sensor data for in situ occupational activity modeling.

Main Methods:

• Collected kinematic data (five postural angles) from 15 male participants performing simulated isometric pushing and pulling tasks.
• Utilized a statistical model incorporating postural angles, anthropometric, and strength measurements.
• Employed a Random Forest algorithm in a two-stage hierarchy for classification.

Main Results:

• The developed model correctly classified 77.2% of handle height (shoulder vs. hip) and force intensity levels (low vs. high).
• Body postural angles were found to be predictable indicators of exerted force magnitude.
• The study demonstrates the potential of wearable sensors for real-time biomechanical assessment.

Conclusions:

• Body posture kinematics derived from wearable inertial sensors can effectively estimate force exertion levels.
• This approach offers a promising method for unobtrusive monitoring of occupational biomechanical exposures.
• Future work can refine these models for broader application in workplace safety and ergonomics.