Concurrent Validity of Two Inertial Measurement Unit Pipelines for Estimating Lumbar and Thoracic Kinematics During...

Area of Science:

• Biomechanics
• Musculoskeletal Health
• Wearable Technology

Background:

• Lifting-related low back pain is linked to lumbosacral and thoracolumbar kinematics.
• Traditional motion capture is lab-bound, limiting real-world biomechanical assessment.
• Inertial measurement units (IMUs) offer potential for naturalistic spine kinematics measurement.

Purpose of the Study:

• To evaluate the concurrent validity of two IMU processing pipelines against optical motion capture (OMC) for lifting tasks.
• To compare a proprietary IMU pipeline (MVN Analyze) with an open-source pipeline (OpenSense) using a validated biomechanical model.

Main Methods:

• Nine healthy participants performed four lifting tasks (stoop, squat, free, asymmetric).
• Simultaneous IMU and OMC data were collected.
• IMU data were processed using MVN Analyze and OpenSense (with OpenSim model).
• Statistical parametric mapping (SPM), Bland-Altman analysis, and Concordance Correlation Coefficients (CCCs) were used for comparison.

Main Results:

• MVN Analyze consistently overestimated spinal flexion-extension compared to OMC.
• OpenSense processing of IMU data showed excellent agreement with OMC-derived kinematics (RMSE ≤ 2.9°).
• Discrepancies were attributed to biomechanical model differences, not sensor limitations.

Conclusions:

• OpenSense, when paired with an appropriate biomechanical model, demonstrates high validity for IMU-based lifting biomechanics assessment.
• XSens IMUs show potential for practical, field-based evaluation of lifting kinematics.
• Accurate assessment may require minimal sensor placement (e.g., chest and pelvis).