AI4Health Free Taster Dataset - COPD Detection

Realistic healthcare data, zero red tape.

Getting access to high-quality healthcare data is slow, expensive, and often locked behind complex governance processes.

That’s why we created the AI4Health Synthetic Data Suite - a fully privacy-safe way to explore and build healthcare AI, instantly.

👉

Why this matters

Researchers, educators, and innovators all face the same challenge:
they need realistic healthcare data to build and test models - but traditional datasets are costly, time-consuming to access, and restricted by strict approvals.

The AI4Health Free Taster Dataset gives you the freedom to start experimenting right away.

What's inside

🧩 Dataset contents
  • 1,000 synthetic patient records
  • 500 COPD-positive and 500 COPD-negative individuals (all aged ≥ 35 years)
  • 12 core variables, including:
  • Demographics (age, sex, region)
  • Lifestyle factors (smoking, physical activity)
  • Environmental exposure (PM₂.₅)
  • Clinical measures (BMI, FEV₁ % predicted)
🧠 Use cases
  • Build and test your first AI model for COPD screening or diagnosis
  • Compare model performance with and without spirometry (FEV₁)
  • Explore interpretability methods such as SHAP to understand model drivers

Why it's safe

This dataset is 100 % synthetic - generated from random seeds using generative AI.
No real patient data were ever used, and there is zero privacy risk.
Yet it reflects realistic patterns and disease relationships seen in real-world studies.

How it helps you succeed

  • Fast: No governance, no waiting - download and start within minutes
  • Flexible: Works with any language or platform (Python, R, Stata, Julia, etc.)
  • Educational: Perfect for classroom teaching or skill development
  • Expandable: Serves as your first step toward full-scale synthetic datasets

Quick start plan

  • Enroll and download the free COPD dataset
  • Load in your preferred environment
  • Run your first classification model (screening and/or diagnosis)
  • Explore your model's behaviour and compare against our benchmarks

What's next

We hope you find this free taster useful.

When you’re ready to go further, choose the edition that suits your goals:
Explore COPD Detection, COPD Exacerbations, Cardiovascular Events, and Type 2 Diabetes Complications.
Created with