One of our teammates suffers from early symptoms of repeated strain injury (RSI), due to his job working as a software programmer. He was the main inspiration for this project. The U.S. Bureau of Labor and Statistics reported that musculoskeletal disorders accounted for 26 percent of all workplace injuries at a cost of $45-60 billion in workers compensation and related costs in 2000. The average number of workdays lost because of RSI is three times the average number of workdays lost for all other types of work-related injuries, and total direct and indirect costs to society were estimated at $1 trillion.
What it does
The system uses machine learning to learn the patient's specific movement patterns. It will match the patient's future movements to these patterns to detect specific tension and stress in the muscle to release it with tailored rehabilitation
Stages of operation: 1. Training phase; 2. Rehabilitation phase; 3. Data transferred to online medical record for monitoring by healthcare professional.
How we built it
We built the backend with MatLab 2016b, imported electromyography (EMG) signals from the Myo. The EMG signals are processed in MatLab with machine learning algorithm to detect features of the patient's specific movements. Following, a specific muscle stimulation rehabilitation is delivered to the patient during the treatment session.
Challenges we ran into
There was an incredible amount of EMG data collected. The processing of that data in MatLab and the making of the muscle stimulation hardware were hurdles to jump through. Ultimately, it is bringing everything together that is the most challenging.
Accomplishments that I'm proud of
We are proud of the ultimate solution we produced for people with RSI.
What I learned
We learned a lot about rehabilitation of RSI.
What's next for RSWhy
We want to incorporate more technology in the rehabilitation system such as functional electrical stimulation to extend the range of services we can provide.