Upload README.md with huggingface_hub

README.md

@@ -1,12 +1,106 @@
----
-title: Radioflow
-emoji: 📉
-colorFrom: red
-colorTo: red
-sdk: gradio
-sdk_version: 6.3.0
-app_file: app.py
-pinned: false
----
-
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# RadioFlow: AI-Powered Radiology Workflow Agent
+
+> **MedGemma Impact Challenge Submission**  
+> Targeting: Main Track + Agentic Workflow Prize
+
+## Overview
+
+RadioFlow is a multi-agent AI system that streamlines radiology workflows by processing chest X-rays through an orchestrated pipeline of specialized agents. Built with Google's Health AI Developer Foundations (HAI-DEF) models.
+
+## Architecture
+
+```
+┌─────────────────────────────────────────────────────────────────┐
+│                    RADIOFLOW ORCHESTRATOR                        │
+├─────────────────────────────────────────────────────────────────┤
+│                                                                  │
+│  ┌──────────────┐    ┌──────────────┐    ┌──────────────┐       │
+│  │   Agent 1    │───▶│   Agent 2    │───▶│   Agent 3    │       │
+│  │ CXR Analyzer │    │   Finding    │    │   Report     │       │
+│  │              │    │ Interpreter  │    │  Generator   │       │
+│  └──────────────┘    └──────────────┘    └──────────────┘       │
+│         │                                        │               │
+│         │                                        ▼               │
+│         │                              ┌──────────────┐          │
+│         │                              │   Agent 4    │          │
+│         └─────────────────────────────▶│   Priority   │          │
+│                                        │    Router    │          │
+│                                        └──────────────┘          │
+└─────────────────────────────────────────────────────────────────┘
+```
+
+## Agents
+
+| Agent | Model | Function |
+|-------|-------|----------|
+| **CXR Analyzer** | CXR Foundation | Process chest X-ray, extract features, detect abnormalities |
+| **Finding Interpreter** | MedGemma | Interpret visual findings into clinical language |
+| **Report Generator** | MedGemma | Create structured radiology report |
+| **Priority Router** | MedGemma | Assess urgency, route to care pathway |
+
+## Quick Start
+
+### 1. Setup Environment
+
+```bash
+# Create virtual environment
+python -m venv venv
+source venv/bin/activate  # On Windows: venv\Scripts\activate
+
+# Install dependencies
+pip install -r requirements.txt
+```
+
+### 2. Configure HuggingFace Access
+
+```bash
+# Login to HuggingFace (required for gated models)
+huggingface-cli login
+```
+
+### 3. Run the Application
+
+```bash
+python app.py
+```
+
+The Gradio interface will launch at `http://localhost:7860`
+
+## Project Structure
+
+```
+radioflow/
+├── app.py                 # Main Gradio application
+├── requirements.txt       # Python dependencies
+├── README.md             # This file
+├── agents/               # Agent implementations
+│   ├── __init__.py
+│   ├── base_agent.py     # Base agent class
+│   ├── cxr_analyzer.py   # Agent 1: CXR Foundation
+│   ├── finding_interpreter.py  # Agent 2: MedGemma
+│   ├── report_generator.py     # Agent 3: MedGemma
+│   └── priority_router.py      # Agent 4: MedGemma
+├── orchestrator/         # Agent coordination
+│   ├── __init__.py
+│   └── workflow.py       # Workflow orchestrator
+├── utils/                # Utilities
+│   ├── __init__.py
+│   ├── visualization.py  # Plotting and overlays
+│   └── metrics.py        # Performance tracking
+└── sample_data/          # Test images
+    └── .gitkeep
+```
+
+## HAI-DEF Models Used
+
+- **CXR Foundation**: [google/cxr-foundation](https://huggingface.co/google/cxr-foundation)
+- **MedGemma**: [google/medgemma-4b-it](https://huggingface.co/google/medgemma-4b-it)
+
+## License
+
+This project is submitted under CC BY 4.0 as required by the competition.
+
+## Acknowledgments
+
+- Google Health AI Developer Foundations team
+- MedGemma Impact Challenge organizers