mou11/clinical-decision-support-agent

🏥 Clinical Decision Support Agent

🎯 Live Demo

👉 Try it on Hugging Face Spaces

🔍 Overview

A production-grade Medical AI Agent that takes patient symptoms as natural language input, searches real PubMed medical literature automatically, and generates structured clinical recommendations with citations. Built with LangGraph for agent orchestration and Groq (LLaMA 3.3 70B) for clinical reasoning.

📊 Results

Metric	Score
BERTScore Precision	0.8449
BERTScore Recall	0.8887
BERTScore F1	0.8663

🧪 Test Cases

Case	Symptoms	Urgency Level
Case 1	Fever, cough, chest pain	High ✅
Case 2	Stiff neck, fever, confusion	Emergency ✅
Case 3	Chest pain radiating to left arm, diabetes	Emergency ✅

✨ Features

• ✅ Real-time PubMed literature search via NCBI API
• ✅ LLM-powered clinical reasoning (LLaMA 3.3 70B via Groq)
• ✅ Structured output: conditions, tests, treatments, urgency
• ✅ Multi-turn conversation memory
• ✅ BERTScore quality evaluation
• ✅ Interactive Gradio web interface

🏗️ Architecture

Patient Symptoms → Extract Search Query (LLM) → Search PubMed (NCBI API) → Generate Recommendation (LLaMA 3.3 70B) → Format Report → Gradio UI

🛠️ Tech Stack

Component	Tool
Agent Orchestration	LangGraph
LLM	Groq — LLaMA 3.3 70B
Medical Literature	PubMed NCBI API
Evaluation	BERTScore
UI	Gradio
Platform	Google Colab (CPU)

🚀 How to Run

1. Get a free Groq API key at console.groq.com
2. Open app.py in Google Colab
3. Replace your-groq-api-key-here with your actual key
4. Run all cells in order

⚠️ Medical Disclaimer

This system is for educational and research purposes only. It does not provide medical advice. Always consult a qualified healthcare professional for medical decisions.

📌 Project Status

✅ PubMed search tool complete ✅ LangGraph agent built ✅ Multi-turn memory working ✅ BERTScore evaluation done ✅ Gradio demo deployed on Hugging Face Spaces