ParcoursVis: Progressive Visualization of Patient Pathways at Scale

Description

The Aviz Team and the French Social Security (CNAMTS) are interested in evaluating patient pathways for treatments. We are currently working on two main projects: (1) ParcoursVis and (2) ParcoursUrge. In ParcoursVis, doctors want to understand the differences between how treatments are de facto given around France and how they align with the official recommendations. In ParcoursUrge, we visualize the events that take place in the Emergency services of the Parisian hospitals to help doctors and hospitals optimize patient management.

Both ParcoursVis and ParcoursUrge projects present several challenges, such as finding methodologies to effectively process large-scale data (multi-million patients with hundreds or thousands of events per patient). They transform raw data collected for prescription reimbursement or monitoring into an effective visualization.

To visualize and interact with this high volume of data, we rely on Progressive Data Analysis. We visualize millions of patient events by aggregating them on the fly by "chunks", groups of 10,000 or so patients, iteratively. At each iteration, we update the visualization that quickly converges to the final result. This method allows aggregating and visualizing billions of patients with no noticeable delays, allowing smooth interactions at any time. We apply it to improve health at the country or regional level, a scale that was impossible to handle before.

This work is funded by the Health-Data-Hub, and AP-HP. The project originally started as a collaboration with CMAP/École Polytechnique and CNAMTS.

Demo

See our demo here. We also provide a user manual. The system is open-source and available at https://gitlab.inria.fr/aviz/parcoursvis.

Contact

E-mail all co-authors

ParcoursVis has been developed with the help of: