Determining inertial measurement unit placement for estimating human trunk sway while standing, walking and running

Area of Science:

• Biomechanics
• Gait Analysis
• Wearable Sensors

Background:

• Inertial measurement units (IMUs) are widely used for estimating medial-lateral (M/L) trunk sway.
• Accurate M/L trunk sway estimation is crucial for assessing and treating gait disorders.
• IMU sensor placement significantly impacts the accuracy of trunk sway estimation.

Purpose of the Study:

• To determine the optimal placement of IMUs for accurate M/L trunk sway estimation.
• To investigate the relationship between multi-segment spine movements and clinically defined M/L trunk sway.
• To develop and validate a trunk sway estimation algorithm using an optimally placed IMU.

Main Methods:

• Ten healthy subjects performed standing and walking trials in a motion capture laboratory.
• Markers were placed on the spine, acromion, and posterior superior iliac spine.
• Multi-segment spine movements were tracked to compare with IMU data and clinical definitions of M/L trunk sway.

Main Results:

• Spine locations T7-T8 and T9-T10 showed the closest match to clinical M/L trunk sway during standing (0.5 deg error) and walking (1.0 deg error), respectively.
• Lower spine locations (L2-L4) demonstrated the least accuracy for both standing (1.5 deg error) and walking (4.0 deg error).
• An algorithm using a single, optimally placed IMU achieved root-mean-square errors of 2.5 deg and peak errors of 2.0 deg during walking.

Conclusions:

• Optimal IMU placement for M/L trunk sway estimation is on the mid-back (T7-T8 or T9-T10).
• This placement reduces errors caused by spinal movements not aligning with the clinical definition of trunk sway.
• Accurate trunk sway measurement using IMUs can enhance the assessment and treatment of gait disorders.