Inspiration

Our inspiration for NeuroBite came from a shared personal experience, each member of our team discovered that we grind our teeth at night. What initially seemed like a minor habit is actually a common and often overlooked health issue linked to stress, sleep quality, and neurological activity.

Bruxism, commonly known as teeth grinding or clenching, often goes undetected and is closely linked to stress, sleep disorders, and neurological activity. Many individuals suffer from dental damage, chronic headaches, and sleep disruption without realizing the root cause.

We aimed to bridge dentistry and neuroscience by creating a solution that not only protects teeth but also provides insights into brain activity and sleep health. We were motivated by research showing the connection between jaw muscle activity and neural systems such as the mesencephalic trigeminal nucleus (MTN) and the ascending reticular activating system (ARAS).

What it does

NeuroBite combines two devices: an EEG headband and a smart dental mouthguard. The EEG detects beta band activity that show when the jaw is about to clench. The smart mouthguard uses pressure sensors to confirm jaw movement. An AI system combines these signals to identify when teeth grinding is about to happen and sends a gentle vibration through the EEG device to stop it. The system also provides sleep and brain activity insights and data through a mobile app.

How we built it

We wrote code in Python to organize and analyze sleep EEG datasets and simulate bite-force data. An AI model was then used to combine both signals and detect upcoming teeth grinding. When grinding is detected, the system sends a gentle vibration through the EEG device to stop it. A simple mobile app was designed to display sleep and bruxism insights.

Challenges we ran into

One of the main challenges was finding suitable datasets for both EEG and bruxism, as there are limited publicly available sources. Synchronizing brain signals with mouth movement data was also difficult because they come from different sensors. Designing a comfortable and realistic mouthguard with embedded electronics was another challenge. Additionally, ensuring that the vibration feedback would stop grinding without waking the user required careful consideration.

Accomplishments that we're proud of

We are proud of creating a unique solution that combines neuroscience and dental health. Our team successfully designed a system where two devices work together to detect and prevent bruxism. We developed code to synthesize and organize datasets and demonstrated how AI can be used to improve detection accuracy. We created a clear concept for a user friendly mobile app that provides meaningful sleep insights and data, with the potential to positively impact society by revolutionizing the dental industry and improving overall human health.

What we learned

Through this project, we learned the importance of interdisciplinary collaboration between healthcare, engineering, and data science. We gained experience in working with EEGs, EEG data, sensor fusion, and AI model decision making. We also learned how to design solutions with the end user in mind, focusing on comfort and usability.

What's next for NeuroBite

Next, we plan to build a fully functional hardware prototype and test it with real users. We aim to improve the AI model using real world data. Collaborating with dentists and sleep specialists will help validate the effectiveness of the system. In the future, we hope to pursue clinical trials and regulatory approval so that NeuroBite can become a widely accessible consumer health product. NeuroBite also plans to grow beyond detecting teeth grinding and tracking sleep. It can be used by dentists and neurologists and may also help monitor stress, sleep disorders, and conditions like epilepsy. The system will connect with wearable devices, telehealth services, and electronic health records, creating a connected platform that supports both users and healthcare providers.

Built With

  • csv
  • dataclasses
  • eeg
  • json
  • logistic-regression
  • pathlib
  • python
  • random-forest
  • sensor-fusion
  • vscode
Share this project:

Updates