AI-Agent Liver Transplant Commitee

📌 Project Overview

This repository contains an AI-driven Liver Transplant Committee Decision System that evaluates patient eligibility for liver transplantation. The system processes clinical vignettes and uses AI agents representing medical professionals—including a Hepatologist, Cardiologist, Transplant Surgeon, and Social Worker—to assess one-year survival probability post-transplant. The project is implemented using CrewAI, LangChain, and OpenAI's GPT-4o for decision-making.

🔍 Features

✅ Multi-Agent Decision Making: Simulates a liver transplant selection committee. ✅ AI-Powered Clinical Vignettes: Generates structured case reports. ✅ Parallel Processing: Optimized patient evaluation for efficiency. ✅ Error Handling: Ensures smooth execution with input validation. ✅ Secure API Integration: Uses environment variables for API keys.

🚀 Installation & Setup

1️⃣ Clone the Repository

git clone https://github.com/your-username/Liver-Transplant-AI-Committee.git
cd Liver-Transplant-AI-Committee

2️⃣ Install Dependencies

Ensure Python 3.8+ is installed, then run:

!pip install -q crewai
!pip install -q crewai[tools]
!pip install crewai[agentops]

3️⃣ Set Up OpenAI API Key

Create a .env file (or set as an environment variable):

OPENAI_API_KEY=your-openai-api-key

🔧 How It Works

1. Load Patient Data: Reads an Excel file containing patient information.
2. Generate Clinical Vignettes: AI processes medical history and generates structured reports.
3. Multi-Agent Committee Review: AI simulates a committee meeting where each agent evaluates the case:

• 🏥 Hepatologist → Assesses liver condition & MELD scores.
• ♥️ Cardiologist → Determines cardiac risks post-transplant.
• 💎 Transplant Surgeon → Evaluates surgical risks & feasibility.
• 🏠 Social Worker → Considers social determinants & post-op support.

4. Final Decision: A consensus is reached, and the output is stored in structured files.