Inspiration

We were captivated about the historic performance of NBA player Stephen Curry when he took the NBA by a storm with this fantastic 3-point shooting. If you've ever played basketball, you probably have yelled "Kobe!" as you attempt to toss paper trash into your garbage can. For enthusiasts of this sport, we introduce a technology that will help elevate your game. Project Form is designed to analyze and optimize your basketball shooting form to perfect your performance in game.

What it does

We developed an iOS smartphone application interfaced with a Myo Armband to analyze and optimize the user’s basketball shooting form. We measure a variety of parameters including arm extension, wrist and elbow deflection. Then, the user’s performance is compared against ideal values determined by the wettest of basketball science. It also provides processed EMG data to monitor the user’s muscle fatigue.

How we built it

The smartphone application was developed in Swift using the Myo Objective-C API. We designed a simple user interface designed for simplicity and ease of use that provides the user with essential information to further improve their shooting form. The Myo Armband is connected to the mobile application via Bluetooth, and is used it to measure a wide range of data metrics such as acceleration, angle of deflection, and EMG data. We utilized signal processing on the EMG data to determine metrics such as wrist deflection, and used acceleration in conjunction with EMG to determine the power output of the user. Additional EMG signal processing was performed by computing the Power Spectral Density of the signals; it has been proven that the decrease in mean or median values of an EMG signal’s power density over time is correlated to muscle fatigue. The user can use this information to prevent muscle injuries caused due to fatigue and it can also be beneficial for diagnostic purposes in the case of a sports injury.

Challenges we ran into

The biggest challenge we experienced was during the development process. Since the Myo API is not available in Swift and most of the team members had very limited knowledge with using Objective C, we had a lot of issues using the Myo API in Swift. Figuring out the optimal way to process the data to provide meaningful information also proved to be a challenge. We also ran into some issues with figuring how to process all the data metrics we wanted. In particular, getting a reasonably accurate calculation for the wrist deflection and processing the EMG data to depict usable information about the muscle’s arm activity.

Accomplishments that we're proud of

We were able to successfully put together a project that integrated software and hardware to teach the users proper basketball shooting technique, which has potential in biomedical applications relating to the study of muscle fatigue. By taking advantage of our research (mainly for EMG signal processing) and problem-solving skills, we compensated for our limited knowledge and still managed to develop a skill-improving and biomedical device.

What we learned

This experience allowed us to explore potential applications of a variety of hardware, but most importantly, it helped us gain more software and hardware knowledge. Specifically, we were able to obtain a deeper insight into EMG signal processing and how to use them for representing meaningful information. Our skills with programming in swift and Objective-C were improved as well.

What's next for Project Form

The next steps for our project include optimizing the acquired data from the Myo Armband to relay more accurate information. We would also like to build upon our current EMG signal processing algorithms to extract more information about activity; for instance, look for symptoms of sports injuries at an earlier stage to prevent and reduce them.

Built With

Share this project:

Updates