Unlocking Health Insights: A Practical Introduction to AI & Data Visualization FHCC 2026

Introduction

Healthcare generates more data than ever, yet much of it remains locked in spreadsheets, out of reach of the clinicians, students, and policymakers who could act on it. This hands-on, half-day workshop demystifies data visualization and artificial intelligence for a primarily non-technical health audience.

Using a real-world (synthetic) clinical dataset in a beginner-friendly, browser-based environment (Google Colab), participants will learn how to transform raw health data into clear, decision-ready visualizations and interpret AI-driven predictions responsibly. Through a guided "code-along" approach with pre-prepared code snippets, the workshop emphasizes concepts and storytelling rather than debugging, enabling participants—regardless of coding background—to build and customize their own health visualizations.

Objectives

• Explain why data visualization matters in clinical and public health decision-making and recognize common pitfalls in health charts.
• Import, inspect, and clean real-world health data, including handling missing values and performing basic data wrangling.
• Build descriptive and exploratory visualizations (distributions and correlations) and progress from static to interactive plots using Plotly.
• Understand conceptually how AI models generate predictions and how to visualize prediction uncertainty and model outcomes using explainability concepts such as SHAP and LIME.
• Independently design and customize a visualization or dashboard that effectively communicates a health insight to stakeholders.

Target Audience

This workshop is designed primarily for clinicians, public health students, researchers, and policy-focused professionals with little or no coding experience (approximately 95% non-technical and 5% tech-savvy participants). No programming background is required. Participants should bring a laptop with internet access to fully participate in the hands-on exercises.

Workshop Structure

The workshop will be delivered as a four-hour, hands-on "code-along" session using Google Colab, requiring no local software installation. The session is organized into the following modules: