Deep Learning-Enhanced EMG Armband with an Interactive Game for Effective Wrist Rehabilitation

Wrist rehabilitation is essential for patients recovering from injuries or surgeries to regain mobility and strength. This study introduces a novel approach using a deep learning-enhanced electromyography (EMG)-based armband integrated with an interactive game. The custom-designed armband captures E...

Full description

Saved in:
Bibliographic Details
Main Authors: Vo, Truong-Tien, Mai, Ngoc-Dau, Choi, Jaeyeop, Oh, Junghwan
Format: Conference Proceeding
Language:English
Subjects:
Accuracy
armband
Continuous wavelet transforms
Deep learning
Electromyography
EMG
fast CWT
Games
Injuries
Muscles
Sensors
Surgery
Wrist
wrist rehabilitation
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Wrist rehabilitation is essential for patients recovering from injuries or surgeries to regain mobility and strength. This study introduces a novel approach using a deep learning-enhanced electromyography (EMG)-based armband integrated with an interactive game. The custom-designed armband captures EMG signals from wrist muscles and processes them with an advanced deep learning model to interpret five wrist movements: power grip, extension, flexion, radial deviation, and ulnar deviation. These movements are translated into game actions, creating an engaging and motivating rehabilitation experience. The game adapts to the patient's progress, offering personalized challenges to promote continuous improvement. Our system achieved classification accuracies of 94.48% for subject-dependent and 91.18% for subject-independent approaches, demonstrating its effectiveness in wrist rehabilitation.
ISSN:2168-9229
DOI:10.1109/SENSORS60989.2024.10785188