A Novel Gait Event Detection Algorithm Using a Thigh-Worn Inertial Measurement Unit and Joint Angle Information

Area of Science:

• Biomechanics
• Robotics
• Wearable Technology

Background:

• Gait event detection is crucial for controlling assistive devices like exoskeletons.
• Existing methods often require multiple sensors or complex setups.
• A simplified, accurate detection system is needed for real-time applications.

Purpose of the Study:

• To develop and evaluate a novel, threshold-based gait event detection algorithm.
• To utilize data from a single thigh inertial measurement unit (IMU) and joint angles.
• To enable detection of heel strike (HS) and toe off (TO) for potential exoskeletal control.

Main Methods:

• Algorithm development using synchronized IMU and ground reaction force (GRF) data from healthy individuals.
• Utilized 3D thigh accelerations and angular velocities.
• Validated timing accuracy and quantified fidelity using F1-Score against GRF data.

Main Results:

• High accuracy for heel strike (HS) detection: mean timing error of -41±20 ms and F1-Score of 0.988.
• Excellent accuracy for toe off (TO) detection: mean timing error of -1.4±21 ms and F1-Score of 0.991.
• Algorithm parameters for TO were identified on a per-subject basis.

Conclusions:

• The developed algorithm demonstrates a promising, simplified approach to inertial-based gait event detection.
• The high F1-Scores indicate strong fidelity in detecting HS and TO.
• Further real-time evaluation and refinement are necessary for practical implementation in exoskeletal control systems.