NicolasGensollen
commented
4 years ago J'ai lu ce papier ce matin sur le thème du Covid 19: https://www.google.com/url?sa=t&source=web&rct=j&url=https://www.epicx-lab.com/uploads/9/6/9/4/9694133/inserm-covid-19_report_lockdown_idf-20200412.pdf&ved=2ahUKEwjIj5HvlOXoAhUrzYUKHVROAjIQFjAAegQIBxAB&usg=AOvVaw0OXJOncCY_TI8hHDugHY6W
Le modèle est assez complexe, mais ça vous donne une idée de ce qui se fait dans le domaine actuellement.
Bon courage!
@ARE2020-G1G2/coronavirus j'ai vu passer un mail qui devrait vous intéresser sur la modélisation du COVID-19.
Voici le lien pour lire les infos et jouer avec leur modèle: lien.
Voici également un copier-coller du mail en question:
At George Mason University, we built during the past few days a preliminary agent-based model of COVID-19 spread. We are updating the version of the model here regularly.
This model is built upon a general spatial model of virus spread, where agents can get infected by being in touch with each other, with a certain probability. This probability is reflected in the "infectiousness" parameter and can be adjusted by the user of the model. We refined this model by tracking the age of each agent in the population and adjusting the chances of recovery based on agent's age (using current COVID-19 data).
We are tracking children with the blue color and the elderly with the yellow color, while the other agents are green if healthy, gray if recovered and potentially immunized, and red if infected. We also included rough climate and humidity exogenous parameters, and adjusted the rate of infectiousness based on this (currently not informed by COVID-19 data). The chances of recovery and infectiousness rates are informed by the COVID-19 data as well, globally, while the birth rates and lifespan of the population are informed by the worldwide real data.
The model can be initialized with any number of agents up to 1000 and, this being a spatial model, we observed that the higher the population density, the longer it takes for the epidemic to die out. We also observed that the number of agents that recover from the infection dips faster but also increases faster in warmer climates, while it deeps slower and increases slower in colder climates. This is ongoing work and we are still to run full parameter sweeps to test for the statistical significance of results. We also plan to include GIS features to the model, so that we allow for realistic spatial definition for the movement of the agents.
You can view and interact with the model here: http://modelingcommons.org/browse/one_model/6227#model_tabs_browse_info
If you are interested, here is some other work some of our external faculty has been working on related to the virus:
SFI External Faculty Caroline Buckee and Marc Lipsitch are co-authors on this research paper:
https://www.medrxiv.org/content/10.1101/2020.02.04.20020495v2
Marc Lipsitch is viewed as an authority on COVID-19 spread at the moment and has been quoted in several articles:
https://www.theatlantic.com/health/archive/2020/02/covid-vaccine/607000/
https://www.npr.org/sections/health-shots/2020/03/04/811146915/how-computer-modeling-of-covid-19s-spread-could-help-fight-the-virus
https://www.nationalgeographic.com/science/2020/02/why-travel-restrictions-are-not-stopping-coronavirus-covid-19/
https://www.medrxiv.org/content/10.1101/2020.02.13.20022707v2
https://www.medrxiv.org/content/10.1101/2020.03.04.20031112v1
Lauren Ancel Meyers has also co-authored papers on the subject:
https://www.medrxiv.org/content/10.1101/2020.02.19.20025452v2
https://www.medrxiv.org/content/10.1101/2020.01.28.20019299v4
Bonne lecture!