BuzzUnfreeze

Inspiration

While volunteering at an elderly home, we met numerous patients affected by Parkinson’s disease, a neurodegenerative disease that impacts the motor system of those affected. We saw first-hand the impacts the disease had on the patients’ quality of life. A common symptom, freezing of gait, puts the elderly at great risk of injury, thus dramatically decreasing their quality of life. We built this device in hopes of countering this condition.

What it does

This device detects freezing of gait in individuals with Parkinson’s disease, via the sound of a buzzer.

How I built it

The algorithm detected the average amplitude of the acceleration in the y-axis and compared it to a threshold value. If it exceeded the threshold value, a signal was sent to the buzzer and a noise was produced, therefore signaling shuffling.

Challenges I ran into

A challenge we ran into occurred when figuring out which acceleration direction worked best to detect shuffling. We tested out the three directions (x, y and z) to determine which provided the most accurate results for our experiment.

Accomplishments that I'm proud of

It was difficult to determine the algorithm for the code, so there was a great sense of pride when we developed a code that accurately performed its function, as well as accounted for any human error. Specifically, it was difficult to account for sudden jerking that could be interpreted as shuffling by the program. To counter this, we decided to calculate an average time value over three periods and compare it to the threshold value, thus increasing the accuracy of the device.

What I learned

Through this project, we familiarized ourselves with the arduino hardware, while expanding our knowledge of the language C/C++. We also learned to apply various concepts from different courses to create a solution to a real world problem (i.e. using acceleration to detect shuffling movement).

What's next for BuzzUnfreeze

We hope to attach a motor to our device so that it can treat the shuffling, as well as detect it. We also hope to further expand the code, so the device can self-calibrate, rather than have its up, down and average threshold values predetermined. In addition, we intend to calculate other values (i.e. maximum acceleration), therefore increasing the accuracy of our device’s ability to detect shuffling.

Built With

• arduino
• c/c++