Elora | Devpost

Waiting for brain signal transmission and analysis.. Showing all signals connected
Neural Network/CNN recording brainwaves
Recognizes emotion and starts generating multiple photos to match emotion, and selects the best one
Loaded stress levels, photo, sentiment, and a full report.
Bottom half
Neural Network Model

Note: The Muse S EEG headset does not provide anything else besides the 4 Channels, TP9 AF7 AF8 TP10

Imagine after a serious traumatic event (PTSD), you are not able to share your emotion and get the proper treatment you need. This is the story for millions of people around the world, increasing DAILY...

Introducing Elora Elora is a brainwave-powered emotion tracking system designed specifically for veterans but really for anyone with PTSD or emotional struggles. Our wearable EEG headset (Muse S Gen 2) detects and analyzes brain activity, using our custom machine learning models to recognize emotions such as stress, anxiety, calmness, and happiness. We then conceptualized this with easy-to-understand graphs and brainwave charts. Not only do we show stress levels and the current emotions you have, but we also show the individual brainwaves, the literal feeling you are experiencing, conceptualized via AI, and and LLM, trained to cater to post-PTSD emotions, which advises on how specialists can treat for you.

Frontend (User Interface):

React.js: The frontend is built with React.js, providing a dynamic and responsive interface for users. It allows us to create a seamless, real-time experience for visualizing the user's emotional data through interactive graphs and brainwave charts. React’s component-based architecture makes it easy to manage the dynamic content of the application, ensuring smooth updates without needing to reload the page. Tailwind CSS: For styling, Tailwind CSS is utilized, enabling us to quickly design a clean and modern user interface. Tailwind's utility-first approach ensures that the design is highly customizable and responsive, providing an intuitive experience for users of all backgrounds.

Backend (Data Processing and AI Models):

Flask: The backend is powered by Flask, a lightweight Python web framework that facilitates easy API integration and smooth handling of server-side logic. Flask is ideal for handling requests from the frontend, especially when dealing with the real-time analysis of brainwave data and emotion tracking.

Elora’s backend is powered by Flask, serving as an API endpoint that processes real-time EEG data from the Muse S Gen 2 headset. The brainwave data—alpha, beta, delta, and theta waves—is collected and stored in .npy format for efficient handling. The deep learning model is a CNN-RNN fusion with LSTM, where CNN layers extract spatial EEG features while LSTM layers capture temporal dependencies, enabling accurate emotion detection. Trained on self-collected data, the model has 50K parameters, balancing efficiency and accuracy for real-time inference. Flask ensures seamless communication, handling incoming EEG data and returning predicted emotional states with minimal latency.

Conceptualization and AI Feedback:

AI-powered emotional guidance: Once the data is collected and analyzed, Elora uses an AI system to conceptualize the user’s emotional state. This AI provides real-time feedback in the form of graphs and charts, helping users understand their feelings better. It also provides emotional advice or coping strategies based on the recognized patterns, guiding users on how to manage their emotions without needing to speak to someone or ask for help.

Developing Elora posed challenges like processing real-time brainwave data, training machine learning models to accurately detect emotions, and creating an intuitive interface that could make complex data understandable. We had to ensure integration of the Muse S Gen 2 EEG headset with our backend and optimize the emotional feedback system to provide personalized, effective advice. Through this hack, we overcame these boundaries to create only the best product for emotional wellbeing.

This is not just a mental health tool; it’s a bridge between science, technology, and emotional resilience, helping veterans understand their emotions without saying a single word. We are proud of the process we've been through, through tears of trial and error

Elora is finished and ready for deployment. Our next step would be partnering with hospitals to address mental health in marginalized communities, to give the best for the world, which is what Elora is all about.