Crisis Care OS

Real time hospital capacity monitoring for emergency response for hospitals and users to navigate.

Comment

CrisisCare OS

Real-time hospital capacity monitoring for emergency response - helping hospitals and users navigate emergency care options.

Project Story

Inspiration

The idea for CrisisCare OS came from observing how difficult it can be to find real-time hospital information during emergencies. When someone needs urgent care, they often don't know which nearby hospitals have capacity or what wait times look like. I wanted to build a tool that empowers both healthcare facilities and patients with transparent, real-time data that could potentially save lives.

What it does

CrisisCare OS is a comprehensive emergency response platform that provides:

  • National Hospital Map - Interactive US map showing 60 real hospitals with live capacity indicators and ER wait times
  • Hospital Hub - Real-time floor view of a hospital showing bed availability, ER status, ICU capacity, and active surgeries
  • Patient Help (JARVIS) - AI-powered assistant that uses voice recognition and camera vision to help users describe symptoms and get guidance
  • Live Dashboard - AI chat interface with text-to-speech for emergency guidance and hospital information
  • Disaster Alerts - Real-time natural disaster tracking with impact assessments

How we built it

The project was built with:

  • Next.js 16 with React 19 for the frontend framework
  • TypeScript for type safety
  • Tailwind CSS for styling with a pixel-art retro aesthetic
  • TensorFlow.js (coco-ssd model) for real-time object detection in JARVIS camera mode
  • Web Speech API for voice recognition and text-to-speech
  • SVG-based US map with interactive hospital markers

Challenges we ran into

  1. Hospital location mapping - Positioning 60 real hospitals correctly on a custom SVG map coordinate system required careful calibration
  2. Voice recognition cleanup - Ensuring AI voice and speech recognition properly stop when navigating away from pages
  3. Balancing real data vs. simulation - Since there's no free real-time hospital capacity API, we used realistic simulated data for capacity/wait times while using real hospital names and locations

Accomplishments that we're proud of

  • Built a fully functional emergency response dashboard with retro pixel-art aesthetic
  • Implemented JARVIS AI assistant with both voice recognition and camera-based symptom analysis
  • Created an interactive national map with 60 real hospital locations across the US
  • Integrated text-to-speech for accessible AI interactions

What we learned

  • How to integrate TensorFlow.js for real-time object detection in the browser
  • Managing complex state with React hooks for multi-page single-application feel
  • Building voice-enabled interfaces with Web Speech API
  • Coordinate mapping between data and SVG visualization

What's next for CrisisCare OS

  • Connect to real hospital APIs (CMS, state health departments) for live capacity data
  • Add mobile app companion
  • Expand JARVIS to handle more complex medical triage scenarios
  • Integrate with emergency services for direct 911 handoff

Tech Stack

First, run the development server:

npm run dev
# or
yarn dev
# or
pnpm dev
# or
bun dev

Open http://localhost:3000 with your browser to see the result. You can start editing the page by modifying app/page.tsx. The page auto-updates as you edit the file. This project uses next/font to automatically optimize and load Geist, a new font family for Vercel.

Learn More

To learn more about Next.js, take a look at the following resources:

Built With

Share this project:

Updates

Submission history