Hello🙏🏻, We are here with ELEVATE
We have all come across depressions and psychological traumas and it's indeed a part of human emotion. During this pandemic, most of us have been isolated at homes and this is where depressions/traumas nurture and sprout! Thus we bring to you ELEVATE... ELEVATE is your one-stop location which is an AI-based depression detection system that assists psychiatrists along with psychologists to guide as well as examine their patients with absolute easiness especially during this time of the pandemic.
What it does
- Individual doctor interface
- Anonymous patient CheckIn/Checkout
- Effective Emotion detection
- Reports submitted to expert psychologists/psychiatrists
How we built it
ELEVATE was created using the following open source projects: *[Tailwind CSS UI models] - Front-End Framework *[HTML]- Front-End Outline
- [Flask] - Python framework for back-end
- [Flask-Socketio] - Flask add on for socket.io support
- [Flask-SQLAlchemy] - Flask package for manipulating databases
- [FlaskWTF Forms] - Flask form handler
- [Keras/Tenserflow] - ML Model
- [OpenCV] - Image/Video Capture
Challenges we ran into
Making sure that we capture the right expressions are able to transfer the data without delay and store it securely are some of the issues we faced. There are some issues that still exist which we are working on like making the data to be analyzed in real-time so that we don't have to store the data as such.
Accomplishments that we're proud of:
We are proud that we were able to develop a web application that resolves prevalent psychological issues and with the proper resources, we are confident that we can make ELEVATE, an ultimate solution for depressions and psychological traumas.
What we learned
We were able to learn and explore several new project ideas and tech stacks. We were able to implement an AI model that detects and analyses human emotion and submits a report based on even micro-emotions prevalent in the face. Through this project, we were able to innovate, ideate and explore several problems prevalent in society, proving that team effort is the best effort!
What's next for Elevate
For now, we have achieved a facial expression recognition model we can actually start working on the AI models for depression detection from text and voice spectrum of the person speaking. For this we would have to conduct a deeper study as analyzing the voice to detect detection makes use of bilateral LSTM's. Once we achieve this we have 3 models that we have to put together to arrive at a conclusion that will decide the state of the patient. Making all these models work in real-time is yet another mystery left for us to crack.
Log in or sign up for Devpost to join the conversation.