Inspiration Our team wanted to create a tool that could genuinely help emergency departments, where nurses juggle large patient volumes and fast decisions. We were inspired by how challenging triage can be under pressure and wanted to build something that keeps things calm on the surface, just like a duck gliding across water. That idea became the heart of Quack and Assess.
What it does Quack and Assess is an AI-assisted triage system that helps nurses quickly identify which patients need attention first. It provides a clean nurse dashboard that displays priority levels, alerts, patient details, and AI-generated key notes. The system analyzes patient symptoms using a weighted scoring model and returns a priority status that updates the dashboard in real time.
How we built it We built the frontend using HTML, CSS, and JavaScript, creating a warm, hospital-friendly dashboard inspired by our logo. The backend uses FastAPI to process requests, run the triage scoring logic, and deliver combined patient and priority data. All patient, priority, and note information is stored in MongoDB Atlas, and our scoring engine uses custom weights defined in our question weights file to determine severity.
Challenges we ran into We faced difficulties with CORS, linking the frontend and backend, matching data models across collections, and getting the dashboard to correctly render live data. Debugging how each part communicated took time, especially when resolving request errors and fixing file structure issues.
Accomplishments that we are proud of We created a fully working triage assistant that connects a functioning dashboard, an AI scoring engine, and a live database. We are proud of how polished and intuitive the interface feels, how accurately the scoring system works, and how quickly we built a complete end-to-end workflow.
What we learned We learned how to integrate FastAPI with MongoDB in a dynamic full-stack environment, how to structure data for clinical use cases, and how important clear communication is between components. We also gained experience designing user interfaces that support healthcare professionals in high-stress situations.
What is next for Quack and Assess We plan to expand the system to include voice-based triage, more advanced AI reasoning, improved analytics, and a mobile-friendly view. Our long term vision is to evolve Quack and Assess into a clinical tool that supports nurses, improves workflow, and helps patients receive timely care
Built With
- cors
- css
- eleven-labs
- fastapi
- google-cloud
- html
- mongodb
- mongodb-atlas
- openai
- python
- scikit-learn
- sql
Log in or sign up for Devpost to join the conversation.