Inspiration
I built this project because many patients leave the hospital without fully understanding their discharge instructions. I wanted to create a system that makes medical information easier to understand and helps patients know exactly what to do after leaving the hospital.
What it does
This project takes patient information like medical conditions and medications, then turns it into a clear discharge summary. It also gives simple instructions for recovery and estimates the patient’s risk of returning to the hospital.
How we built it
We built the system as a simple AI agent with three main parts.
It uses FHIR to get patient information, a risk-scoring tool to estimate readmission risk, and an AI agent layer to generate clear discharge instructions.
All parts work together to turn raw medical data into simple, understandable output for both doctors and patients.
Challenges we ran into
One of the main challenges was working with medical data in a way that is both accurate and easy to understand. It was also challenging to design the system so that different parts like data retrieval, risk scoring, and the AI agent could work together smoothly.
Accomplishments that we're proud of
We successfully built an AI system that can turn real patient data into clear and useful discharge instructions. The system also estimates patient risk and explains medical information in a way that is easy for non-medical users to understand.
What we learned
We learned how to work with healthcare data using the FHIR standard, how to design simple AI systems that combine different tools, and how important it is to make technical information easy for patients to understand.
What's next for Discharge Planning Agent
Next, we plan to improve the system by adding more advanced AI models to make the discharge instructions even more accurate and personalized. We also want to connect it to real hospital systems and support more patient conditions and use cases.

Log in or sign up for Devpost to join the conversation.