Hypersensitivity of trunk biomechanical model predictions to errors in image-based kinematics when using fully...

Area of Science:

• Biomechanics
• Spine Research
• Medical Imaging

Background:

• Advancements in medical imaging enable displacement-control trunk models for estimating spinal loads without muscle force calculations.
• The sensitivity of these models to post-imaging displacement measurement errors remains uninvestigated.

Purpose of the Study:

• To assess the sensitivity of musculoskeletal (MS) and finite element (FE) spine models to errors in measured vertebral displacements.
• To determine the impact of translation and rotation errors on biomechanical model predictions.

Main Methods:

• A Monte Carlo analysis was employed to evaluate model sensitivity.
• Six static activities across different postures were simulated using a hybrid MS-FE model.
• Computed displacements drove displacement-control MS and FE models under varying error levels (0.1-0.3 mm translations, 0.2-0.6° rotations).

Main Results:

• Both MS and FE models exhibited substantial task-dependent sensitivity to vertebral translation errors.
• Intradiscal pressures (IDPs) were significantly affected (SD up to 1.05 MPa) with increasing translation errors.
• Axial compression and shear forces showed direction reversals with translation errors of 0.3 mm.
• Model outputs were less sensitive to rotational errors.

Conclusions:

• Current vertebral translation errors (≥0.1 mm) in image-based kinematics are too large for accurately driving spine biomechanical models.
• The findings highlight critical limitations for using displacement-control models in clinical applications without improved measurement accuracy.