Update README.md

README.md

@@ -1,47 +1,11 @@
-# # CMS Coding Changes Mining for RCM 🏥
-
-Building a forward-looking analytics platform to anticipate and interpret CMS regulatory changes using Agentic AI and Machine Learning.
-
-## 🚀 Key Features
-
-- **Executive Intelligence Scorecard**: Real-time simulation of revenue impact from 250+ CMS rule changes with 12-month projections.
-- **AI-Driven CDM Auto-Sync**: Batch auditors scan **2,500+ records** in seconds to prevent $2.45M+ in billing denials with automated file persistence and backups.
-- **Predictive Denial AI**: High-accuracy risk scoring matching legacy billing codes against 2025 "Packaging" rules.
-- **Advanced Financial Projections**: Full fiscal year impact modeling with automated 'Net Impact' trend analytics.
-- **Agentic Reasoner**: Optimized 5-agent LangGraph system (Regulatory, Finance, CDI, CDM, Orchestrator) with executive-ready insights.
-- **Billing-Ready Assets**: One-click, multi-sheet Excel export with full audit trails and data dictionaries.
-
-## 📂 Project Structure
-
-- `app.py`: Streamlit frontend with interactive dashboards, simulation lab, and AI advisor.
-- `ml_engine.py`: Core Intelligence Engine containing Random Forest models, impact simulation logic, and CDM patching automation.
-- `agent_graph.py`: LangGraph implementation for multi-agent collaborative reasoning.
-- `data_generator.py`: High-fidelity synthetic data generation for claims, rules, and CDM.
-- `data/`: Persistent storage for claims, rules, CDM, and FAISS vector index.
-
-## 🛠️ Getting Started
-
-1. **Install dependencies**:
-   ```bash
-   pip install -r requirements.txt
-   ```
-2. **Generate base data** (Initializes FAISS index & training datasets):
-   ```bash
-   python data_generator.py
-   ```
-3. **Run the application**:
-   ```bash
-   streamlit run app.py
-   ```
-
-## 🧠 AI & Machine Learning Architecture
-
-The system utilizes a hybrid AI/ML approach:
-
-- **Predictive ML**: Random Forest Classifier for denial risk prediction based on categorical and numerical claim features.
-- **Agentic AI**: A directed acyclic graph (DAG) of specialized agents exploring regulatory data using RAG (OpenAI + FAISS).
-- **Automation Engine**: Algorithmic auditing and patching for Chargemaster (CDM) integrity.
-
----
-
-_Temple Health Unified Intelligence Hub | Deployment Edition_
+---
+title: TemHealth
+emoji: 🏥
+colorFrom: blue
+colorTo: green
+sdk: streamlit
+sdk_version: "1.32.0"
+python_version: "3.10"
+app_file: app.py
+pinned: false
+---