Note: The data used to produce this app is confidential, so only the R code is uploaded to my GitHub account.
This was one of the first R Shiny applications I’ve ever made, and it holds a special place in my heart, not only because of that, but also because it was made in preparation for the first Symposium I’ve ever presented at: the 2017 UCI UROP Symposium!
For the 2017 UROP Symposium, I decided that it would be a cool idea to make a clearer representation of the 3-dimensional scatterplot I originally made in R with the scatterplot3d package after I noticed that it (which can be found in the poster linked to at the bottom of this post) had the data points all clumped together, making it very difficult to see any correlation or pattern. The 3-dimensional scatterplot I made here (with the rgl package) displays a regression line and can be rotated and zoomed in and out, which makes it much easier for the observer to get an idea of how the data is behaving.
Each data point is the 5-year risk percentage of a woman developing and being diagnosed with breast cancer, at the intersection of the three most widely recognized and validated breast cancer risk assessment models: the Gail, Tyrer-Cuzick, and BCSC. Different colorations of the data points represent different ethnicities of women. Pink data points represent Hispanic women, purple represent White women, and orange represent Asian women.