Validation of the Perception Neuron system for full-body motion capture

Area of Science:

• Biomechanics
• Human Motion Analysis
• Wearable Technology

Background:

• Inertial Measurement Units (IMUs) offer a portable alternative to traditional optoelectronic motion capture systems.
• The Perception Neuron system (PNS) is a commercially available IMU for motion analysis.
• Previous validation of PNS was limited to upper body, single-plane movements.

Purpose of the Study:

• To further validate the Perception Neuron system (PNS) against a VICON optoelectronic system.
• To assess PNS accuracy during dynamic, full-body movements including walking, jogging, and sports-specific actions.
• To determine the feasibility of PNS for full-body kinematic analysis.

Main Methods:

• Comparison of PNS with VICON motion capture system.
• Analysis of dynamic movements: walking, jogging, and multi-articular sports actions.
• Validation metrics included Pearson's R correlation, Root Mean Square Error (RMSE), and Bland-Altman analysis.

Main Results:

• Most joint angles showed a Root Mean Square Error (RMSE) below 4° and a strong average Pearson's R correlation of 0.85.
• Shoulder abduction/adduction exhibited higher RMSE and moderate correlation.
• Bland-Altman analysis indicated a mean bias of less than 10° for most joint angles, with exceptions in shoulder and elbow movements.

Conclusions:

• The Perception Neuron system (PNS) demonstrates good performance against VICON for full-body kinematic analysis.
• PNS may not be ideal for replicating precise raw joint angles.
• PNS is suitable for detecting changes in movement patterns and comparing normalized joint angles.