Inspiration

We were inspired by the idea that AI can simplify one of the most frustrating parts of healthcare: the referral process. Today, referrals are often delayed due to missing information and poor communication between providers and patients. We saw an opportunity to use AI to automate these gaps, reduce delays, and make the process easier to understand for patients.

What it does

VitaSync is an AI-powered medical referral assistant that streamlines the referral process from start to finish. It analyzes referral information in real time, detects missing documents, and guides both patients and providers on next steps. It also translates complex medical and administrative details into clear, user-friendly language, giving patients a structured view of their care journey—from referral submission to specialist appointment.

How we built it

We built VitaSync using a modular architecture: Frontend: React for an interactive user interface Backend: FastAPI to manage API requests and workflows AI Layer: Fetch.ai agent to analyze referrals and generate structured outputs The system processes clinical notes, identifies the appropriate specialty, detects missing information, and generates task lists and patient-friendly explanations that prepare patients for their appointments. All components communicate through APIs, making the system scalable and easy to extend.

Challenges we ran into

One of the main challenges was handling unstructured medical data and turning it into reliable, structured outputs. Medical notes can vary widely in format and clarity. Another challenge was balancing technical accuracy with simplicity—ensuring that explanations were medically correct while still being easy for patients to understand. We also had to design a workflow that works for both providers and patients without adding extra complexity.

Accomplishments that we're proud of

We’re proud that we built a system that not only automates referral analysis but also improves patient understanding. Building an encompassing app, we wanted to address all the challenges that patients face when trying to make an appointment. This includes real-time detection of missing referral information, automatic generation of personalized patient task lists, translation of medical jargon into clear, actionable steps, and working end-to-end prototype connecting frontend, backend, and AI.

What we learned

We learned how powerful AI can be in reducing administrative burden in healthcare, especially when applied to real-world workflows. We also gained experience in designing systems that handle unstructured data, building modular, scalable architectures, and creating user experiences that prioritize clarity and accessibility.

What's next for VitaSync

Next, we plan to integrate with electronic health record (EHR) systems, add automated appointment scheduling, expand multilingual support for broader accessibility, and improve compliance with healthcare data standards and security. Our goal is to make VitaSync a fully integrated tool that reduces delays, improves coordination, and makes the healthcare journey clearer for every patient.

Built With

  • agentchatprotocol
  • asi-1
  • fastapi
  • fetch.ai
  • figma
  • ngrok
  • python
  • react
  • rest
Share this project:

Updates