Seizure Sensor

Judging Category: Best Beginner App

Inspiration

Epilepsy and seizure-related deaths and injuries have always been a concern for millions of people around the world, which inspired us to try to create this app. By easily incorporating this already common technology (Fitbits) in their daily lives, medical professionals and close ones could efficiently respond to seizures, ultimately saving lives.

What it does

The Fitbit is worn by patients throughout their everyday lives, and after downloading the app, the accelerometer feature would track their movement and use the data to predict potential seizures. If the patient experiences rapid shaking/spasms, the device would notify emergency contacts/medical professionals.

How I built it

I used Fitbit Studio and coded in JavaScript (and a little HTML and CSS) so that the seizure status that is displayed on the sensor is based on the accelerometer data (if it was a distance of 10 in all directions). The accelerometer uses orthogonal technology in 3 axes/directions. In addition, we coded the background color and icon of the app so it is visually pleasing.

Challenges I ran into

First of all, we were originally going to train data and use ML but we couldn't find relevant data sets for accelerometer statistics and seizures. Also, when it came to the messaging, we realized that Fitbit actually does not support sending automated messages to external devices, so we spent a lot of time and effort on trying to find a way around that, but in the end, we realized that it would not be possible in this timeframe (we could continue working on this later).

Accomplishments that I'm proud of

I'm proud that our simulations of seizures actually result in a response from the Fitbit (including the vibration). In addition, I am pleased that the Fitbit does not sense normal/stable movement (like walking and running) as a seizure.

What I learned

I learned a lot about using hardware to solve real-life problems. In addition, I learned how to code in JavaScript, HTML, and CSS, and incorporate data into software.

What's next for Seizure Sensor

There is a lot that we hope to do with Seizure Sensor! We are considering using Apple Watch because of its much more useful messaging apps and how it can easily connect to other people (this is our next step that we need to incorporate to potentially save someone's life). In addition, we are considering implementing Machine Learning so that our sensor is much more accurate.