The French Defense Procurement Agency (DGA) and the Defense Innovation Agency (AID) are funding solutions based on artificial intelligence to organize the response to the Covid-19 epidemic. Among the projects selected: ONADAP * software. This platform, developed by the Borelli Center and the Army Training Hospital (HIA) Percy, aims to quickly visualize the health situation of nursing staff on a hospital scale and to anticipate the spread of the virus within it.
Visualization and decision support software
Concretely, this visualization and decision support software is divided into three parts. The first one, the ‘personal section’, must be developed in the short term. This will allow managers to view the health status of staff in real time and to plan for three days (with any feedback from people on arrest).
The ‘patient section’, developed in the medium term, will make it possible to assess in real time the saturation rate of the service dedicated to patients affected by Covid-19 and to detect any services through which these patients may have passed before being taken care of. Finally, the ‘simulation and decision support component’, developed over the long term, aims to provide a tool for modeling the spread of Covid-19 within the health structure and to make comparisons between the different strategies for allocating human resources and equipment.
Lots of data
This project, which began on April 24, must continue over an 18-month period. The algorithms are developed by the Borelli Center and the software by ITNovem, a subsidiary of the SNCF. The modeling part is structured around three phases: the integration of predictive models, the definition of propagation models, and the experimentation of these models which will take place over the last six months of the project’s duration.
For these modeling tools to work properly, a lot of data must be collected, such as the number of staff and patients, those who have Covid-19 and those who are suspected of having it, their function, the date of appearance symptoms, etc. In order to feed the mathematical models, the partners therefore conduct a seroepidemiological study at the HIA Percy, the aim of which is notably to determine the number of people affected by Covid-19.
* The acronym ONADAP stands for: Digital Decision Support Tool for dynamic strategies for allocating human and material resources during the Attrition Period for Staff from health units or forces in the Covid-19 epidemic