FOCUS ON OUR FACULTY:
A professor with a vision for visualization
Dr. Daniel B. Carr
Professor of StatisticsDan’s expertise is in the field of statistical graphics and visual analytics. He is enthusiastic about quantitative graphics because it allows him to collaborate with researchers and students in many different fields. Dan is currently writing a book with Linda Pickle on visualizing patterns in data with micromaps.
Dan took a sabbatical at the National Cancer Institute to work with Linda Pickle and Sue Bell in developing graphics for the planned statecancerprofiles web site. The web site went live in 2003 and NCI uses this usability assessed site to communicate with health planners across the nation. Figure 1a shows a screen shot of a linked micromap plot from http://statecancerprofiles.cancer.gov/micromaps. Figure 1b shows the printer friendly output from the web site. Both show lung cancer mortality rates and a risk factor, an indicator of cigarette smoking rates. Heath planners seek to reduce preventable causes of cancer death through prevention and early detection. The green area in the left column of dots plots is the Healthy People 2010 target for reducing lung cancer mortality. The green area in the right column of dot plots is the Healthy People 2010 target for reducing cigarette smoking. Health planners can see at a glance from the dots showing rates and confidence intervals how their state is doing relative target area.
Click on picture for original image
Figure 1a: National Cancer Institute's State Cancer Profiles Web Site Example: Lung and Bronchus Cancer
Click on picture for original image
Figure 1b: The printer friendly output for Figure 1a showing all the states without scrolling.Dan initially developed this design to address Tony Olsen’s (US EPA) challenge to provide geospatial context for simple statistical graphics. Most of the early linked micromaps examples focused on environmental applications. Later examples addressed labor and health statistics.
Dan seeks challenges and tries to push back boundaries. While at the National Cancer, Myong-Hee Sung provided the challenge to visualize statistics about the docking of peptides (segments of proteins) on human immune system molecules. The docking of peptides indicates the cell has been infected and should be destroyed. When natural human peptides dock and trigger the immune system, the results are autoimmune diseases such as arthritis and lupus. When foreign peptides fail to dock the human body is subject to attack without defense. One hope in the study of docking statistics is the development of vaccines that would promote the docking of peptides from HIV proteins.
