NCT07222085

Long-Term Stability of the Glide Control Strategy

Determining the Long-Term Stability of the Glide Control Strategy for Upper Limb Prostheses

• Status: Recruiting
• Phase: N/A
• Study type: Interventional
• Enrollment: 12 (estimated)
• Sponsor: Infinite Biomedical Technologies · Industry
• Sex: All
• Age: 18 Years
• Healthy volunteers: Not accepted

Summary

This research is intended to test whether the prescription of the Glide prosthesis control system reduces the burden of use for both patients and their clinical care team as compared to use of Pattern Recognition-based advanced myoelectric control. The goal of the study is to fill the gaps in clinically relevant knowledge to inform the prescription of prosthesis components and the rehabilitation process.

Conditions

• Upper Limb Amputation Above Elbow
• Upper Limb Amputation Below Elbow

Interventions

Type	Name	Description
DEVICE	Glide Control System	Glide is a commercially developed directional myoelectric control strategy from Infinite Biomedical Technologies (IBT) that sits between classic Direct Control (DC) and modern Pattern Recognition (PR). Instead of requiring an isolated on/off muscle signal for each function (e.g., DC) or training a complex classifier on many gestures (e.g., PR), Glide uses the relative activity across 2-8 EMG electrodes to move a virtual cursor on a 2-D "Glide map." The map is divided into adjustable sectors ("slices"), and each slice is assigned a prosthetic movement (e.g., hand open/close, wrist rotation, elbow flexion). Moving the cursor into a slice actuates that movement.
DEVICE	Pattern Recognition System	Pattern recognition (PR)-based myoelectric control is a data-driven approach that allows a user to control multiple prosthetic functions using natural muscle activation patterns rather than discrete, isolated signals. Instead of mapping one muscle to one motion (as in conventional Direct Control), PR systems record the spatial and temporal pattern of EMG activity from multiple sites on the residual limb and use machine learning algorithms to classify which intended movement the user is trying to make.

Timeline

• Start date: 2025-10-15
• Primary completion: 2028-11-01
• Completion: 2029-04-01
• First posted: 2025-10-29
• Last updated: 2025-11-05

Locations

10 sites across 1 country: United States

Regulatory

• FDA-regulated device study