By: (Team 43) Esther Usen, Jack Simonton, John Anvik, Julie Wojtiw-Quo, Kelvin Ying, Steven Gu
For natHACKS 2024 focused on synthesizing EEG data.
Intro • How to Use • Future • Contributors • Links • Resources
Current neuroscience research studies involve real-time EEG collection, pairing recorded brain activity with specific images, actions, emotions, or spoken words. However, the scarcity of accessible EEG datasets limits broader research - gathering data can be costly and time-consuming. To tackle this problem, we reversed the typical data flow by training machine learning models to generate EEG data.
Run: python main.py
- Improve upon our SAAS by generating other types of EEG data corresponding to images, emotions, spoken words and additional motor functions.
- Add GPU capabilities to the model predictions, which would reduce our prediction/inference time of 5-10 seconds down to milliseconds.
- Fine tune the model weights and layers in order to make the output data even more realistic to real-world, live captured EEG data.
- Add website / online capabilities and a server-hosted backend so that users do not need to download any software
- Steven Gu - ML Specialist
- Kelvin Ying - Data Scientist
- Julie Wojtiw-Quo - Project Manager, Data Scientist
- John Anvik - Backend, Frontend
- Jack Simonton - Full Stack
- Esther Usen - Frontend, Backend