For example, the age range 18-64 is pretty broad, especially when considering "above 60" is high risk.
One recent article from The Journal of allergy and clinical immunology, 12 April 2020, "Risk factors for severity and mortality in adult COVID-19 inpatients in Wuhan" splits it up from 0-44yrs, 44-64, >65. That study is looking at adults specifically, but additional population age customization would allow for modeling with more varied hospitalization rates as age increases.
The other population improvement that would be nice is somewhere, maybe next to the graph, an input for the total population. This would allow a more intuitive look at the data because most people don't think in "per 100,000" numbers. If the population of the U.S. is 340ish million people, and I see 250 million people infected, that could be more impactful.
I ran into these issues when trying to validate SimPandemic against data available from JHU for Italy.
Otherwise, a very interesting tool!