Community /

Texas Engineering World Health: Low-cost Wearable Seizure Prediction Alarm with Cyton and Ultracortex Mark IV

This is an update from the OpenBCI Discovery Program. Click here for details on how to apply.

1. What are you making?

 Texas Engineering World Health seeks to design and implement a low-cost seizure prediction device that can be easily accessible for epileptic patients throughout the world.  We aim to develop a  medical device capable of predicting the onset of a seizure by using a machine-learning algorithm to analyze data from a brain-computer interfacing EEG headset. The desired outcome of the device is to be able to predict epileptic seizures 1-5 minutes before the onset and alert the user via an alarm/notification system.  After we have developed a working prototype for seizure prediction, we will add a protective layer to the headset through CAD (Fusion 360). Modeled after the RibCap, an existing headgear that protects users from seizures and other neurological conditions, our design will enhance both comfort and safety.     

Fig 1. Sketch of preictal data flow; Raspberry Pi 3 outputs to alarm system 

2. How are OpenBCI tools being applied?

We plan to pair our machine learning model with the 16 channel 10-20 system Ultracortex Mark IV which will alert the user and their family of an onset of a seizure. The electrode system will be connected to an Open BCI Cyton+Daisy board that filters the raw EEG signals. These signals will be sent, via Bluetooth, to a Raspberry Pi, which will hold a machine learning model. If the model detects the presence of preictal waveforms in sensor data, which indicates the onset of a seizure, then the Pi will flash red lights, and make a loud, ambulance-like sound. This would signify to surrounding pedestrians that the user needs help. 

3. Why is this important?

One of the most dangerous consequences of epilepsy is the possibility of sudden unexpected death in epilepsy (SUDEP). Considering that patients were found to be prone and pulseless just 15 minutes after the beginning of a SUDEP-inducing seizure, it is believed that devices and methodologies allowing for quick or premature intervention before each seizure is necessary for epilepsy patients to avoid potential SUDEP or general bodily injury.  As opposed to existing ventures which focus on seizure detection, our project is focused on seizure prediction. This simple solution can help prevent death from SUDEP-inducing seizures.

4. Who is involved in this project?

This project is being completed by the 2020-2021 Seizure Detection Team under Texas Engineering World Health (TEWH). TEWH, an organization under the University of Texas at Austin, works on medical projects for the developing world. Each year we work on a different project and submit it to the Engineering World Health Competition. We work on researching the prevalent problems that need solutions and we look into making the idea financially feasible for hospitals in the developing world. Lastly, we create a prototype that encompasses engineering skills from all different engineering fields. The general members of our organization range from freshman- seniors and are typically one of the following majors: Electrical, Mechanical, Chemical, and/or Biomedical engineering. This diverse membership has proven useful in past projects and allows for greater collaborations and skills to be brought to the team. You can learn more about us here on the TEWH website. 

We thank OpenBCI for admitting us into the sponsorship program, and we look forward to updating the community on our progress!

Leave a Reply