We are taking an open data set (Cook County case/fatality stats & hospital information) and mixing it with a simulated piece of information (hospital ICU beds occupied) to provide a dashboard that simulates what might be useful for COVID information.
NOTE: The hospital ICU occupancy is simulated data and not accurate. All other data is real and pulled from publicly available sources.
We are using Sympatic’s VirtualVault® to import & normalize the data. The data is being shown in multiple layers that show the example of how the resulting analysis can provide insight (ward, precinct, community area, and census tract totals) without revealing the source data (shown as fatality locations/demographic data).
The flow is: data is pulled, normalized, dropped back into S3, and then analysis is run & produces the above visualization. In both steps, the VirtualVault® is destroyed, leaving only the resulting output, showing how the data can be used for impact and protected by only leaving the original sources. The normalized data can be persistent, and in the case of an actual owner, would be left with that owner to manage as they see fit.
The hospital simulated data set is calculated by using the zip code data to approximate the ICU bed utilization, matching the code of zip code cases to a bell curve with the average ICU bed usage for Chicago.