Scienctists develop new tool to help in the fight against the epidemic
A team of Polish scientists have been working on a system for modelling the COVID-19 epidemic which can be used to create restriction scenarios and prepare management strategies at the city, county or commune level.
Developed using current medical knowledge, the system entitled ProME (Prognostic Modelling of the COVID-19 Epidemic) can adapt to the variables of viral biology meaning it can be used now and in the future.
The essence of the project is to create solutions that can be easily adapted to various epidemic conditions and scientists working on the project are constantly updating knowledge about the biology of the virus.
Data science expert and project leader Professor Marek Niezgódka said: “Our goal is to create a tool that will assist the government in making the best and most effective decisions from the point of view of a quick fight against the epidemic at the lowest social cost.”
The team from Cardinal Stefan Wyszyński University in Warsaw (Center for Digital Science and Technology), University of Warsaw, University of Warmia and Mazury and a group of outstanding Polish doctors participating in the creation of models during the pandemic, developed three models.
Two of them (a predictive model and a simulation model) can support the management of the epidemic in Poland. The third model will support health care management processes.
The scientists said: “ProME enables the preparation of management strategies at the city, county or even commune level. Forecasts are made on the basis of real data or specific analogies. For example: we have real data for county X. The system looks for the similarity of epidemic curves and searches for a sequence in a given model that is analogous to the one currently studied. On this basis, it prepares a forecast for county Y.”
Professor Andrzej Szałas from the University of Warsaw, responsible for creating models in the project, added: “First, we prepare the data for analysis. Then, the model is trained, evaluated and delivered in the production version. From that point now on, we monitor the quality of the model and check whether the calculations do not differ from reality.”
If the model starts to generate errors in predictions or calculations, scientists update and clean up the data. As a result, the model learns anew and adapts to the current reality. This process should be performed approximately every two weeks.
According to the scientists, the use of ProME in the fight against a pandemic is a way to precisely implement restrictions on a micro scale.
Professor Niezgódka said: “A regional approach to the restrictions could not only affect the effective fight against the pandemic, but also translate into economic and social benefits.
“It is not necessary to close all sectors of the economy throughout the country, it can be done selectively and locally, thus increasing the effectiveness of the implemented solutions and, perhaps most importantly, social trust in this difficult time.”
The team is currently working on creating a model that will assist the health service in relocating medical procedures to restore the pre-COVID-19 possibility of carrying out medical treatments as soon as possible. (PAP)
ekr/ agt/ kap/