Overall design and implementation of the virtual glove

Area of Science:

• Biomedical Engineering
• Rehabilitation Technology
• Computer Vision

Background:

• Hand rehabilitation is crucial for patients with conditions like stroke, with recovery depending on therapy frequency and quality.
• Existing mechanical rehabilitation gloves are costly, bulky, and not reusable, limiting their application for personalized therapy and data collection.
• There is a need for accessible, accurate, and reusable tools to monitor hand mobility and guide rehabilitation.

Purpose of the Study:

• To design and implement a virtual glove system for accurate, low-cost, and non-bulky hand movement tracking.
• To develop a system that facilitates rehabilitation and evaluates patient progress through precise hand mobility data.
• To present the overall design and implementation of the virtual glove in both real-time and off-line modalities.

Main Methods:

• Utilizes a set of surrounding video cameras to capture synchronized videos of the patient's hand.
• Employs a numerical hand model to track movements and calculate physical, geometrical, and mechanical parameters.
• Implements a marker-based hand tracking algorithm for off-line analysis, including marker positioning, coloring, labeling, detection, and classification.

Main Results:

• The virtual glove system accurately tracks hand movements and calculates relevant parameters.
• Experimental measurements from model-based hand tracking were reported and compared with real hand poses.
• An error estimation strategy was developed and applied to the collected measurements.

Conclusions:

• The virtual glove offers an accurate, low-cost, non-bulky, and reusable alternative to traditional mechanical rehabilitation gloves.
• The system provides a viable solution for patient-specific hand mobility assessment and rehabilitation guidance.
• Further research will focus on system limitations and future improvements for enhanced functionality.