SecureCare

Inspiration

I wanted to make healthcare documentation faster and safer using AI, preventing human errors in prior authorization.

What it does

SecureCare automates prior authorization requests by analyzing clinician notes, detecting sensitive data, and generating structured packets securely.

How we built it

I built it with FastAPI for the backend, React + Tailwind for the UI, and integrated AWS Bedrock for AI-driven analysis and workflow generation.

Challenges we ran into

Getting Bedrock authentication right and parsing structured outputs from large language models took a lot of debugging and iteration.

Accomplishments that we're proud of

I'm proud that SecureCare runs end-to-end: users can input text or files, the system detects key medical data, and can determine whether the users intent is malicious or not.

What we learned

I learned how to design secure AI systems that handle sensitive data responsibly and how to balance usability with strong backend reliability.

What's next for SecureCare

I plan to expand SecureCare so it can automatically fill out the medical portion of actual prior authorization forms instead of displaying what needs to be filled out.

Built With

• aws-bedrock
• boto3
• fastapi
• openai-api-(optional)
• pypdf2
• python
• react
• reportlab
• tailwind-css
• typescript
• vite