ChatGBD:
AI Agent Platform
for Global Burden of Disease Analysis

An intelligent, agentic system for analyzing and visualizing GBD-style epidemiological data through natural language interaction. Powered by LLMs and advanced data science tools.

Get Started See Demo
Python 3.11+ FastAPI Streamlit LangChain
chat_gbd.py 
# Natural language query
user_query = "What's the trend of cardiovascular DALYs in MENA from 1990–2020?"

# Agent processes and answers
agent.answer_question(user_query)
# → Returns analysis + visualization
# → "Cardiovascular DALYs increased from X to Y (+Z%)"
# → Interactive trend plot generated

Powerful Features

Everything you need for intelligent epidemiological data analysis

🤖

Autonomous Agents

Multi-agent architecture with specialized agents for data cleaning, exploration, visualization, and insights.

💬

Natural Language Interface

Ask questions in plain English. No SQL or complex queries needed. The agent understands and responds.

📊

Automated EDA

Intelligent exploratory data analysis with automatic detection of patterns, anomalies, and insights.

🧹

Data Harmonization

Automatic detection and correction of data inconsistencies, missing values, and format variations.

📈

Interactive Visualizations

Publication-quality charts: trends, choropleths, treemaps, and regional comparisons using Plotly.

🔄

Agentic Autonomy

Agents reason, plan, and execute Python code autonomously to answer complex analytical questions.

See It In Action

Real examples of what ChatGBD can do

User Query:
"What are the top 5 causes of death in Eastern Mediterranean Region in 2019?"
Agent Response:

Analysis: Based on the 2019 data, the top 5 causes of death in the Eastern Mediterranean Region are:

  1. Cardiovascular diseases: 31,000 deaths
  2. Ischemic heart disease: 28,500 deaths
  3. Respiratory infections: 12,000 deaths
  4. Neoplasms: 18,500 deaths
  5. Lower respiratory infections: 9,200 deaths

The agent generated a horizontal bar chart visualizing these results and can provide detailed breakdowns by country if requested.

📊 Interactive Bar Chart (Top 5 Causes)
User Query:
"Show me the trend of cardiovascular DALYs from 1990 to 2020"
Agent Response:

Trend Analysis: Cardiovascular DALYs have shown the following pattern from 1990 to 2020:

  • 1990: 45,000 DALYs
  • 2000: 49,000 DALYs (+8.9% increase)
  • 2010: 51,000 DALYs (+13.3% total increase)
  • 2020: 53,000 DALYs (+17.8% total increase)

The agent automatically generated a time-series plot showing the upward trend, with the option to break down by country or compare with other regions.

📈 Interactive Trend Line Chart (1990-2020)
User Query:
"Compare cardiovascular disease DALYs between Iran and Syria from 2000-2020"
Agent Response:

Comparative Analysis:

Year Iran DALYs Syria DALYs Difference
2000 49,000 29,000 +20,000 (Iran)
2010 51,000 30,000 +21,000 (Iran)
2020 53,000 31,000 +22,000 (Iran)

The agent generated a comparative visualization showing both countries' trends side-by-side, with Iran consistently showing higher cardiovascular DALY burden throughout the period.

📊 Comparative Line Chart (Iran vs Syria)

System Architecture

Built with modern, extensible technologies

Frontend
Streamlit UI
↓
Agent Layer
Data Agent Insight Agent Viz Agent
↓
LLM Framework
LangChain
↓
Data Processing
pandas numpy plotly
↓
API Backend
FastAPI

Core Modules

  • Data Ingestion: CSV/Excel upload with schema validation
  • Data Cleaning Agent: Autonomous preprocessing and harmonization
  • Insight Agent: Natural language question answering with code generation
  • Visualization Agent: Publication-quality chart generation
  • Exploration Agent: Automated EDA and pattern detection

Get Started in Minutes

Quick setup and you're ready to analyze GBD data

1

Install Dependencies

pip install -r requirements.txt
2

Generate Example Data

cd data && python generate_example_data.py
3

Launch Application

streamlit run app.py
📖

Full Documentation

Read the comprehensive README
⚡

Quick Start Guide

5-minute setup tutorial
🔧

API Reference

FastAPI endpoint documentation