Using Glove Sensory for Data Acquisition and Processing of Hand Movements

RM 207

The most common source of adult disability in the United States is from a stroke with nearly a hundred thousand attacks occurring yearly. The aftermath of strokes can cause spasticity, which are severe cramps or spasms muscles. Those subjected to long durations of painful muscle contractions experience retrogression with their bodily coordination, limiting their movement or locking limb positions. Therapies, which help reduce spasticity and regain control include range of motion exercises, stretching, frequent repositioning of limbs, and others. Research have shown that technologicallyassisted hand therapy helps regain hand coordination in subjects, encourages increase of at-home exercises, has the potential to be low-cost and be used for generalizing hand gestures. Although past studies exhibited successes, there is still plenty of room for improvement towards using wearable sensors for spasticity therapy. The goal of this project is to implement Virtual Motion Labs’ Data Glove Lite (VMGLite) to collect spatial sensory input of hand positions and develop algorithms for gesture recognition with applications towards physical rehabilitation. This project relies on VMGLite’s advanced spatial capabilities to sense the percentage of fingers bending and the overall hand movement along the roll, pitch, and yaw. This will require software, such as MATLAB and VMGLite SDK, where the latter is programmed on C++ and is editable on Visual Studios and runs through VMGLite Manager. Upon the data acquisition and calculations for gesture recognition, the algorithm is to be complimented with other devices or software to create immersive hand therapy exercises for individuals recovering from strokes.