We believe that Seisir is a revolutionary and life-saving technology. Since traditional seizure and epilepsy monitoring technologies are limited to capturing data over relatively brief periods of time, which can generally be measured most commonly in minutes, occasionally in days and rarely in weeks. With Seisir data can consistently be captured for days, weeks, months or even years without interruption. Not only does this expanded timeframe give Seisir the opportunity to become extremely valuable from a clinical and diagnostic standpoint. Seisir can monitor individuals routinely during periods of change during sleep wake cycles, a time at which seizures are more likely to occur. It is our hope that Seisir can provide a new benchmark in the tools to combat SUDEP, seizures and epilepsy.

What it does

Seisir pulls data from a Microsoft Band, through the app, and then uploads the information to a secure server. Once an individual's data is on the server, Seisir takes advantage of Microsoft Azure to analyze data in real-time using an advanced machine learning algorithm. The program then contacts one's family members and medical staff by way of text messaging in cases where the machine learning program detects a high likelihood of seizure activity. In plain terms, Seisir has a strong potential to save lives by notifying others in emergency cases.

How we built it

Our program in it's entirety utilizes a combination of technologies starting with the Microsoft Band. A bluetooth connection to our android application is established with the band to obtain a variety of sensor data. This information is then passed through the android application to Microsoft Azure on the cloud. Using Microsoft Azure's machine learning platform, our program identifies seizure activity from the sensory data originally derived from the Microsoft Band. The cloud system responds to the android app in moments when seizure activity is found. Upon receiving a seizure alert from Microsoft Azure, the android app notifies emergency contacts via a traditional text message.

Challenges we ran into

Since the Seizure platform requires integrating a combination of multiple different technologies, devices and programming languages. This cross compatibility created the majority of our implementation challenges.

Accomplishments that we're proud of

We successfully developed a machine learning algorithm using Microsoft Azure that can detect seizure simulations with 100% specificity and 100% sensitivity. We understand that our hack uses a limited dataset, but we are excited by the initially promising results.

What we learned

Our team learned a great deal about Microsoft products. We believe that Seisir demonstrates that Microsoft Band and Microsoft Azure have significant clinical applications.

What's next for Seisir - Epilepsy Monitoring App

It is our hope that the Microsoft technologies behind Seisir have the opportunity to be used in a research setting that is clinically relevant to seizures and other epileptiform activity. Since one of our team members interns within the neurology department at Boston Children's Hospital and works with epilepsy patients on a daily basis, it is our hope that the Seisir / Microsoft combination platform have the opportunity to be tested within a clinical setting.

Share this project: