ICUBAM an application to monitor and visualize in real time the availability of intensive care beds

In my daily research work, I work with many resuscitators on issues of causal inference and treatment effectiveness. Quite naturally at the beginning of the health crisis in France, I was in contact with my network of resuscitators and I was notably put in contact with Antoine Kimmoun, a resuscitator from the East of France.  The Eastern region was heavily impacted by the epidemic and the capacity to receive mechanically ventilated patients in intensive care was quickly saturated. The medical players in the field therefore identified the urgent need to view the available COVID+ beds in real time. Antoine thus began developing a prototype for information collection by resuscitators on March 18, 2020 to circulate information throughout the region. We exchanged on this subject on March 21st and on March 22nd we started to develop the open source project ICUBAM with a team of engineers and researchers from Polytechnique, Inria and open source contributors, working full time on the project. The first version of ICUBAM was launched on March 25th.  The application allows a resuscitation network to inform in less than 15 seconds the number of occupied and available COVID+ beds, the number of entries and exits and to have precise information. ICUBAM was originally designed as a field operational tool to assist resuscitators. ICUBAM is used in the east of France by 100 resuscitators and 40 departments (95% of resuscitation units) and has been deployed throughout France with more than 40 departments, 130 resuscitation units, and more than 2,000 resuscitation beds. ICUBAM was very quickly adopted by resuscitators.
ICUBAM is a team effort that has benefited from the work of many contributions, including Gabriel Dulac-Arnold and Olivier Teboul (open-source contributors) who have been involved since the beginning of the adventure.

The data collected by ICUBAM (stored at Inria) are valuable for measuring the burden on the resuscitation services during the Covid-19 pandemic. Current efforts are therefore focused on using the data to predict future bed requirements in real time and more globally for a better understanding and modelling of the epidemic. It should also be noted that the number of resuscitation patients is an indicator of the hospital system's ability to adapt by increasing its capacity.

ICUBAM is intended to be a tool that can be easily deployed in times of crisis in an emergency context, so we hope to stop using it as soon as possible! However, we remain mobilized in this period of deconfinement because the visualizations available in ICUBAM are useful to report on the situation.

We are also helping to deploy the tool abroad. Better planning of resource needs (here the allocation of resuscitation beds) and better triage of patients can have a substantial impact on patient care. We hope that ICUBAM can be useful as an effective decision support tool by collecting detailed and reliable data.

The objective of ICUBAM is both societal, we hope to help patients and different professionals as well as scientific through data analysis.

 

ICUBAM, designed in record time, allowed resuscitators in the field to know which beds were available in real time. Thus, the most serious patients were discharged without loss of time and without causing saturation of the resuscitation units. Subsequently, epidemiological data guided the adjustment of the available beds to handle Covid-19 patients.

Antoine Kimmoun, Doctor in the resuscitation department of the CHRU Nancy

The success of ICUBAM is, in my opinion, linked to the interdisciplinary nature of the team, which includes engineers, researchers, and emergency physicians with different skills (from biostatistics, machine learning, statistical physics, computer science, and resuscitation) who together designed and adapted the entire process, from data collection to real-time analysis and communication of results, to meet operational needs in an emergency context. Data quality is a major challenge in managing this crisis, and flexibility in data collection and direct contact with all stakeholders to obtain and use important information is essential.

ICUBAM is the innovative and easy-to-use application that allows you to have live all the availability of resuscitation beds at regional and national level. Essential for managing the Covid-19 crisis.

Alexandre Boyer, Intensivist doctor and resuscitator at Bordeaux University Hospital

This new project, ICUBAM, led by Julie Josse, professor at the École polytechnique and member of the XPOP Inria research team, demonstrates the impact of research on societal issues, and the importance of combining expertise to find innovative solutions to current problems. The ICUBAM solution developed with Inria helps doctors by monitoring the availability of resuscitation beds and modeling patient flows, thus enabling them to anticipate needs, all through a scientific approach that efficiently processes heterogeneous data from various sources, on different time and space scales. With Bruno Sportisse, we are very pleased that researchers from X and Inria are working closely together on innovative scientific projects.

Eric Labaye, President of Ecole polytechnique & President of Institut Polytechnique de Paris