University of Washington Logo

Panic in the Streets

Data visualization for food-borne disease outbreaks

client

University of Washington

Role

Data Visualization Designer

focus

Interactive Data Visualization

scope

Research, Design, Prototyping

Problem Statement

How might we communicate various factors like ingredients, location and species related to food-borne outbreak in various states of United States to help users find possible relations and trends?

Key User Tasks

  • Track outbreak trends over time
  • Identify most harmful contaminants
  • Analyze high-risk food preparation locations
  • Compare incident control effectiveness by state
  • Identify common species/serotypes by region
  • Track genotype mortality rates
  • Analyze state-wise outbreak vulnerability

Data Processing

Data Acquisition

Dataset obtained from CDC's Food-borne Disease Outbreak Surveillance System (1998-2015)

Data Profiling

Categorized data types and improved quality using Open Refine and Excel

Data Preparation

Extracted relevant categories and standardized nominal data

Data Processing Overview

Design Process

Initial Sketches

Initial concept sketches and ideation

Final Concept

Final concept development

Design Sheet 2
Design Sheet 3
Design Sheet 4
Design Sheet 5

Usability Testing

Testing Approach

Our usability testing process followed a two-phase approach to validate and refine the visualization design. We started with quick guerilla testing using paper prototypes, then progressed to more structured testing with high-fidelity prototypes.

Phase 1: Paper Prototyping

  • Quick iterations with low-fidelity mockups
  • Early feedback on layout and interactions
  • Rapid design modifications based on user input
Paper prototype testing showing map visualizations and data charts

Phase 2: High-Fidelity Testing

  • Detailed interaction testing with working prototypes
  • Validation of visual design and data presentation
  • Final refinements to improve user experience
High-fidelity prototype testing with interactive elements

Design Principles

Shneiderman's Information Seeking Mantra

  • Overview: Storyline approach for high-level understanding
  • Relate: Comparison of illnesses and hospitalizations
  • Filter: Interactive customization for user needs
  • Details on demand: Drill-down capabilities

Future Work

  • Integration with live CDC datasets
  • Incorporation of preventive measures and reporting mechanisms
  • Enhanced data handling with improved infrastructure
  • Additional detailed views of specific outbreaks
Back to Portfolio