Danny T.Y. Wu, Derek Shu, Khanh Le, Ruthik Abbu, Kai Zheng
Abstract: Understanding clinical workflow is a crucial first step to improve the quality, safety, and efficiency of patient care delivery. It enables quality improvement processes and provides a basis to compare and quantify workflow improvements. A common source of data for studying clinical workflow is through time and motion studies, which generates multi-dimensional datasets that are challenging to analyze. Visual analytics can be an effective technique to show patterns and bottlenecks in the time and motion data. Moreover, workflow analysis often involves mixed-method design. The triangulation between the quantitative and qualitative data would require the support of a powerful data exploration tool. To address these challenges, we applied visual analytics to develop a clinical workflow analysis tool (CWAT) that allowed for easy identification of significant workflow patterns. In this system demonstration paper, we describe the visualization design choices and validation through case studies