Development of a Visual Analytics Tool for Polytrauma Patients: Proof of Concept for a New Assessment Tool Using a Multiple Layer Sankey Diagram in a Single-Center Database.

INTRODUCTION: Early physiological assessment of multiple injured patients is crucial for decision making and has relied on personal experience of trauma experts. We have developed a new visual analytics tool (Sankey diagram, Watson Trauma Health care tool) that includes known prognostic parameters for polytrauma patients to help guide assessment and treatment decisions for physicians involved in trauma care. METHODS: A prospectively collected trauma database of a single level I trauma center (3655 patients) was used. INCLUSION CRITERIA: age >16 years, an injury severity score (ISS) >16 and presence of a complete data set in the database. Data collected included admission values of patient age, injury scoring, shock classification, temperature, acid-base and hemostasis parameters. All of these parameters were collected daily as longitudinal parameters. Endpoints of the clinical course we considered were sepsis, SIRS and early in hospital mortality (<72 h). A proof of concept of the visualization was developed over a 2-year period in a cooperation between physicians and engineers. Statistically, the most predictive parameters were selected by binary logistic regression and ROC analysis. RESULTS: A dynamic interactive multilayer Sankey diagram, based on cohort similarities, was developed in a collaboration between the University Hospital of Zurich, Department of Trauma and IBM, from August 2017 until January 2018. It is a modular tool and allows any user to add a new patient, or work with an existing case. The visualization used the data-driven documents (D3) interactive visualization library to create a responsive graphic. CONCLUSIONS: This application summarizes the experience of 3655 polytrauma patients and might serve as a guide for clinical decisions and educative purposes, as well as new scientific questions for the polytrauma patient. LEVEL OF EVIDENCE: IV.

A Data-Driven Decision-Support Tool for Population Health Policies.

Epidemiological models are key tools in assessing intervention policies for population health management. Statistical models, fitted with survey or health system data, can be combined with lab and field studies to provide reliable predictions of future population-level disease dynamics distributions and the effects of interventions. All too often, however, the end result of epidemiological modeling and cost-effectiveness studies is in the form of a report or journal paper. These are inherently limited in their coverage of locations, policy options, and derived outcome measures. Here, we describe a tool to support population health policy planning. The tool allows users to explore simulations of various policies, to view and compare interventions spanning multiple variables, time points, and locations. The design's modular architecture, and data representation separate the modeling methods, the outcome measures calculations, and the visualizations, making each component easily replaceable. These advantages make it extremely versatile and suitable for multiple uses.