Inertial Sensor-to-Segment Calibration for Accurate 3D Joint Angle Calculation for Use in OpenSim

Area of Science:

• Biomechanics
• Human Movement Analysis
• Wearable Technology

Background:

• Inertial capture (InCap) systems combined with musculoskeletal (MSK) models are promising for ecological 3D joint kinematics.
• Accurate sensor-to-segment calibration is a critical limitation for estimating body segment orientations.

Purpose of the Study:

• To develop and validate a novel integrated sensor-to-segment calibration method for InCap systems used with MSK models.
• To compare the accuracy of InCap-derived 3D joint kinematics using the new method against gold-standard 3D motion capture (MoCap).

Main Methods:

• Eleven healthy subjects performed walking, running, and stair ascent/descent trials wearing Xsens InCap and Vicon MoCap systems.
• A novel integrated calibration method was developed and applied to an MSK model with 3 DOF hip and knee joints.
• Root Mean Square Error (RMSE), Range of Motion (ROM), peak values, and R-squared were calculated and compared between InCap and MoCap.

Main Results:

• The integrated calibration method reduced RMSE for hip and knee joints below 5 degrees.
• No statistically significant differences were observed between InCap and MoCap kinematics across all analyzed movements.
• The developed method demonstrated improved sensor-to-segment calibration accuracy compared to MoCap.

Conclusions:

• The novel integrated calibration method enhances the accuracy of 3D joint kinematics estimation using InCap systems with MSK models.
• This approach provides a clinically applicable and easy-to-use solution for accurate human movement analysis.
• InCap systems, with this improved calibration, offer a viable alternative to MoCap in ecological settings.