Two GE Infographic Designs Among Fast Company’s Best of 2011

January 3, 2012

Last week, Fast Company magazine picked the 22 best infographic designs of 2011. Two of them were from GE. Health InfoScape looked at associations between various health conditions, and the Innovation Barometer ranked countries based on their approach to innovation.

Infographics are a smart way to analyze and make sense of the billions of gigabytes of digital data generated every year.

Health InfoScape was the result of a partnership between GE’s data visualization team and computer scientists from MIT’s SENSEable City Lab. The teams crunched 7.2 million anonymized patient records from GE’s proprietary database to investigate the relationships between various health conditions. For example, when you have heartburn, do you also feel nauseous? Or if you’re experiencing insomnia, do you tend to put on a few pounds, or more? As Fast Company’s Cliff Kuang described it, the graphic’s approach to understanding how different ailments are connected could help turn every M.D. into TV’s famously non-intuitive physician, Dr. House.

The researchers reasoned that looking at the data this new way could help us illustrate general health trends and provide new insights about prevention and the healtcare system in general.

The second data visualization chosen by Fast Company was the Innovation Barometer. Measuring innovation is a difficult task. Innovation is a loose concept that can be gauged in many different ways. GE partnered with the graphic designer Lisa Strausfeld and created an infographic based on tangible “innovation drivers” such as education, efficiency, and patents. The graphic also looked at how business executives ranked their countries on innovation, and measured countries on innovation outcomes in jobs, healthcare, environment, and well-being.

Check out GE Data Viz Team’s interactive infographic that compares real innovation drivers and outcomes to executive perceptions of them in twelve countries, using data from GE’s Innovation Barometer.

Last year GE also partnered with Seed Media Group and launched Visualizing.org, an open community of data designers and enthusiasts. One of the group’s key efforts is a series of some 40 contests that challenge participants to boil down a flood of information from a broad variety of sources and transform it into a comprehensive graphic that tells a compelling story. One example of such challenge was the Visualizing Marathon 2011.


This entry was posted in Graphics, Other. Bookmark the permalink.
  • DanielSomerville

    Are you kidding ?

    The Health InfoScape is unreadable by the ~15% of the population who have deficient color vision.
    Even for others, to have to search for 16 different colors is ridiculous.

    This graphic is about as useful as an Ishihara color-blindness test; which is what it appears to be at first glance.

    Maybe Edward Tufte’s book:The Visual Display of Quantitative Information should be required reading for these “artists”.

  • Preston Kemp

    I must agree with Daniel’s comments. These graphics are maybe not in the outright “chart junk” category, but they do need improvement, particularly elimination of the 16-color legend. All of Edward Tufte’s books should be required reading for anyone attempting clear visual communication of data (His one-day class is excellent).

  • Judith P. Oppenheim

    Yes, it is odd that so many of the categories have shades of blue, green or purple while the red, orange, and yellow are used to each represent one category. It would be nice if one could drill down into the graphic and get all 16 categories if one was familiar with the information. To start with, probably a legend with no more than 7 (ROYGBIV – colors of the rainbow) should be used.

  • Brian Mickle

    As a member of the 15% I must agree with the above comments. If a color is not in the Crayola Box of 8 it should not be used in a chart intended for public use. I am constantly having to coach my team on the need for a high contrast pallete when they make charts. A combination of shapes and colors would be easier to read in detail.

  • Jayme Bulhoes

    I can understand your concerns regarding the color limitations.
    But if you play a little with the tool, it is very easy to use, very intuitive. We can magnify, pass mouse over each cluster to read its details, look each gender separately (I was amazed the way the correlation changes when we change from male to female!). You can magnify and, keeping mouse left button pressed, shift graph around. Can also select a specific category to look indivitually at its clusters.
    It is a huge task to extract correlations from so many variables that may interfere with each other. I believe this tool realy made a great contribution to this task.

  • http://profile.yahoo.com/LSQYNFEKLXQHANUW4X5ILY7KRM Bob Stanford

    As far I am concerned I think the Health Infoscape fulfill perfectly its goal of simplying the existing relationship of a complex set of data in a user centered dynamic way and gave me a good opinion of GE.  This visualization could benefit a lot of other community of school projects and it would be great if GE/MIT Media Lab could share the code. Thanks