Ludwig-Maximilians-Universität and Alexander Thamm GmbH work on early warning system for new corona infections

from | 3 June 2020 | [at] News

Alexander Thamm GmbH and Ludwig-Maximilians-Universität München (LMU) are working together on a project to better combat Corona. The aim is to predict daily updated infection figures ("nowcasting") so that regional health authorities in particular can take appropriate measures against further spread of the virus at an early stage or relax existing restrictions. The method will also optimise the data-based information flow for authorities and could also be used in other medical areas in the future.

Together with the Munich Ludwig Maximilian University starts the Alexander Thamm GmbH (AT) is running a project for the data-based fight against the Corona pandemic. The Munich-based data science and AI provider is supporting the "nowcasting model" developed by LMU by adding machine learning and deep learning methods. Professor Dr Göran Kauermann, Dean of the Faculty of Mathematics, Computer Science and Statistics at LMU, explains: "The goal of the joint project is to provide local authorities and health departments with statistically processed information and valid predictions about infections on the ground, as well as to automate the flow of information to them." Decision-makers will thus gain far-reaching insight into the local infection situation, right up to an early warning system.

Without an all-encompassing view, it is a great challenge for the individual institutions to correctly classify the pandemic and derive the right measures from it. Current figures only show the number of newly infected persons in the respective area of responsibility of the public health department. This figure always lags behind the current infection incidence. Reliable projections into the present and especially the future are possible with the help of statistical models and procedures (so-called nowcasting), which, however, are based on national data and must be broken down to the individual districts by statistical extrapolation. In addition, statistical uncertainties are difficult to determine due to different test frequencies and the interpretation of the available data is sometimes complex. Especially with increasing new infections, as expected in autumn, this results in planning and control uncertainties.

Optimised information flow through data science

That is why the data science provider and the LMU want to use "nowcasting" to make accurate estimates of the daily infection incidence and expand them into short- and medium-term forecasts. In the process, data on confirmed infections but especially deaths provide conclusions about the number of new infections. "Although it may sound cynical, the number of deaths is statistically more informative about the number of people actually infected than the number of reported infections, because it does not depend on different testing strategies, accuracies or availability of tests," says Dr Ursula Berger from the Institute of Biometry and Epidemiology, LMU. The head of the LMU Statistical Consulting Laboratory, Professor Dr Helmut Küchenhoff, adds: "Our model already predicts the number of new infections much better than other methods." 

Integration of Deep Learning to minimise statistical uncertainties

In order to be able to develop tools that support health authorities or other local institutions in assessing the situation and selecting appropriate measures, various processes are to be integrated. These include:

  • Data collection and management, including consideration of the data end-to-end process
  • Further statistical modelling and merging of the models, including simulations and modelling of different future scenarios.
  • Complementary modelling with deep learning methods
  • Provision and communication of information

The project improves end-to-end data management, creates transparency at an early stage and enables targeted infection control," explains Andreas Gillhuber, CO-CEO and project manager on the part of Alexander Thamm GmbH. "That's why we see it as a valuable tool in the fight against the current pandemic, but also against other infectious diseases such as influenza or the norovirus." 

About Alexander Thamm GmbH:

The data & AI consultancy Alexander Thamm GmbH is a leader in the development and implementation of data-driven innovations and business models in German-speaking countries. The service portfolio covers the entire data journey - from the data strategy to the development of algorithms and the construction of IT architectures to maintenance and operation. The company's own Data Academy offers training in data science, big data and artificial intelligence. The academy was foundedhe Alexander Thamm GmbH in 2012 by Alexander Thamm and currently employs over 150 people. The headquarters are located in Munich. Other locations are Berlin, Frankfurt, Leipzig, Stuttgart and Cologne. Its clients include more than half of the DAX 30 companies.
You can find more information at Network with us on LinkedIn: 

Press contact:
Michaela Tiedemann
Chief Marketing Officer
Tel: +49 176/1891 7438


Michaela Tiedemann

Michaela Tiedemann has been part of the Alexander Thamm GmbH team since the early start-up days. She has actively shaped the development from a fast-moving, spontaneous start-up to a successful company. With the founding of her own family, a whole new chapter began for Michaela Tiedemann at the same time. Hanging up her job, however, was out of the question for the new mother. Instead, she developed a strategy to reconcile her job as Chief Marketing Officer with her role as a mother.

0 Kommentare