Effect of optimization criterion on spinal force estimates during asymmetric lifting

Area of Science:

• Biomechanics
• Ergonomics
• Occupational Health

Background:

• Optimization-based muscle force prediction models are integral to lumbar spine research and ergonomic assessments.
• These models typically function as subroutines within job analysis software.
• Various optimization criteria exist for determining muscle forces, including minimizing cubed muscle intensities and employing double linear programming.

Purpose of the Study:

• To compare the predictions of two distinct optimization-based muscle force models.
• To assess if these models yield significantly different estimates of spinal forces.
• Specifically investigate differences during a dynamic asymmetric lift task.

Main Methods:

• A laboratory study was designed to evaluate two optimization criteria for muscle force prediction.
• The models compared were: sum of cubed muscle contraction intensities and double linear programming (minimizing spinal compression and maximum muscle intensity).
• Measurements were taken during a dynamic asymmetric lifting task.

Main Results:

• Statistically significant differences were observed between the predictions of the two models.
• Despite statistical significance, the difference in peak spinal compression force was minimal, at only 1.1%.
• This suggests a high degree of convergence in spinal force estimation between the models for this specific task.

Conclusions:

• While distinct optimization criteria lead to statistically different muscle force predictions, the impact on peak spinal compression force is negligible for dynamic asymmetric lifts.
• These findings support the use of either model for estimating spinal loading in ergonomic practice, with minimal impact on key outcome measures.
• Further research could explore other dynamic tasks or different populations to confirm generalizability.