Healthcare big data is being widely touted as a potential resource for curbing costs and improving outcomes. However, numerous challenges remain for leveraging this data to its full potential. In this position paper, we identify the difficulties that characterize clinical data, based on our experiences working with pediatric asthma data from Children's Healthcare of Atlanta. The specific dataset we explored includes administrative items, medications, lab results, clinical respiratory scores (outcome), timestamps, and demographic information from 5,785 emergency department (ED) visits for asthma exacerbations. We argue that new data and visual analytic techniques are needed that are specifically tailored for solving challenges in healthcare, and we propose characteristics that these techniques should have and give our design rationale. To demonstrate how a tool that embodies these desirable features may be designed, we introduce CareProcessVis, a prototype interactive visual analytics tool that helps clinicians explore and understand the processes involved in pediatric asthma emergency department care.
The Computational Enterprise Science Lab focuses on the design, analysis, and management of complex enterprise systems (e.g. organizations, supply chains, business ecosystems) using information visualization, modeling/simulation, and system science approaches.