State estimation of walking phase and functional electrical stimulation by wearable device

Area of Science:

• Biomedical Engineering
• Rehabilitation Technology
• Neuroscience

Background:

• Functional electrical stimulation (FES) aids gait improvement in patients with peroneal nerve palsy or post-stroke spastic hemiparesis.
• Accurate phase switching detection is crucial for effective FES application in walking.
• Existing methods may lack the precision required for real-time FES control.

Purpose of the Study:

• To develop and validate a wearable device for estimating walking state and delivering functional electrical stimulation (FES).
• To implement and train an artificial neural network (ANN) for precise phase switching detection in gait.
• To assess the practical applicability of the developed device for FES timing.

Main Methods:

• Design of a wearable device integrating sensors for gait state estimation.
• Implementation of an artificial neural network (ANN) for supervised learning of gait phase transitions.
• Conducting two experimental trials to evaluate the device's performance in estimating FES timing.

Main Results:

• The wearable device successfully estimated walking states and gait phase switching timings.
• The artificial neural network (ANN) demonstrated effective supervised learning for precise timing.
• Experimental results confirmed the accuracy of the FES timing estimation.

Conclusions:

• The developed wearable device is effective for state estimation and functional electrical stimulation (FES) timing.
• The proposed ANN-based method provides accurate gait phase switching detection for practical FES application.
• This technology holds promise for enhancing rehabilitation outcomes in individuals with gait impairments.