NamaSleep

Inspiration

Sleep health is often overlooked, yet it directly impacts long-term well-being. We wanted to build a system that detects risks early and helps both patients and doctors take action before conditions worsen.

What it does

NamaSleep analyzes patient lifestyle and sleep data to generate a risk score, flag high-risk cases, and provide personalized recommendations. It also gives doctors a dashboard to monitor patients, prioritize care, and refer them to specialists when needed.

How we built it

We built a full-stack app using FastAPI for the backend and Next.js for the frontend. We used SQLite + SQLAlchemy for data storage, created a custom risk-scoring engine, and integrated AI-generated summaries for both patients and doctors. We tried integrating Tailwind to accommodate frontend of the AI recommendations.

Accomplishments that we're proud of

We successfully built an end-to-end system with real-time analysis, doctor feedback loops, and a working referral system, all within a short time. The platform is both functional and scalable.

What we learned

We learned how to design full-stack systems quickly, manage databases properly, and build AI-assisted features that are actually useful and interpretable in a healthcare context.

What's next for NamaSleep

We plan to incorporate larger datasets to improve model validity, make AI feedback more personalized, and further train our RL model with more data and episodes. We’ll add analytical tools for doctors to track patient trends, improve UI and coding practices, integrate wearable devices for real-time data, and implement secure patient/doctor authentication.

Our submission is entered in the Artificial Intelligence and Human Computer Interaction challenges.