Skip to content

AMNETIZEN/claude_hackathon

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

2 Commits
 
 

Repository files navigation

claude_hackathon

🏥 ClearCare Triage

Empowering patients through clear information and ethical AI boundaries.

Submission for the Anthropic Hackathon (IIT Delhi, April 12, 2026) - Track 1: Biology & Physical Health


📖 Overview

ClearCare Triage is a Streamlit-based health literacy platform designed to translate complex medical jargon and lab reports into accessible, 5th-grade reading-level English. To protect patient safety, the platform features a strict Ethical Handoff Protocol that bypasses the AI and redirects users to immediate emergency care if high-risk symptoms are detected.

Healthcare is expensive, inaccessible, and confusing. Medical knowledge that could help people is locked behind jargon. ClearCare aims to bridge this gap, ensuring that patients understand their own health data without crossing the dangerous line into AI medical diagnosis.


🎯 The Core Question

"Who are you building this for, and what could go wrong?"

We are building this for underserved patients who are overwhelmed by medical jargon and struggle to navigate the healthcare system. The biggest danger in health-tech is an AI providing false reassurance for a fatal condition (e.g., dismissing chest pain as heartburn).

We mitigated this risk by hardcoding a keyword-based Handoff Protocol. The moment severe risk factors are detected, the AI is immediately locked, and the user is forcefully redirected to a human medical professional.


✨ Key Features

  • Jargon Translator: Users can paste confusing text from lab reports or doctor's notes, and the AI translates it into plain, comforting English.
  • Symptom Triage: A chat-like interface for users to describe how they are feeling to get educational context about their symptoms.
  • ⚠️ The Ethical Handoff Protocol: The core safeguard of the platform. If the system detects critical keywords (e.g., "chest pain," "shortness of breath," "stroke"), it triggers an emergency UI takeover, stopping all AI processing and directing the user to the nearest emergency room.

🛠️ Tech Stack

  • Language: Python
  • Frontend & Framework: Streamlit
  • Core Logic: Rule-based keyword extraction layered with AI translation simulation.

🚀 How to Run Locally

  1. Clone the repository:
    git clone [https://github.com/your-username/clearcare-triage.git](https://github.com/your-username/clearcare-triage.git)
    cd clearcare-triage

About

No description, website, or topics provided.

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

No releases published

Packages

 
 
 

Contributors