Monitoring of disease outbreaks (MOOD)

Monitoring disease outbreaks in the context of data science

Challenge

The detection of the occurrence of infectious diseases is usually based on the reporting of cases, i.e. on so-called indicator-based surveillance (IBS). However, this method lacks sensitivity, as cases are often not reported or are reported late. In an environment that is changing due to climate change, animal and human mobility, population growth and urbanization, there is an increased risk of the emergence of new and exotic pathogens that may go undetected in IBS. Therefore, there is a need to detect signals of disease occurrence using informal, multiple sources, i.e. event-based surveillance (EBS). The MOOD (MOnitoring Outbreak events for Disease surveillance in a data science context) project aims to harness data mining and analysis techniques for big data from various sources to improve the detection, monitoring and assessment of emerging diseases in Europe. To this end, the MOOD project is developing an analysis and visualization platform that enables real-time analysis and interpretation of epidemiological and genetic data in combination with environmental and socio-economic variables in an integrated cross-sectoral, interdisciplinary One Health approach:

Data mining methods for collecting and combining heterogeneous big data (geo, satellite, weather and social media data),
A network of disease experts to identify the causes of disease occurrence,
Data analysis methods applied to this big data to model the occurrence and spread of disease vectors and infectious diseases,
an online platform for end-users, i.e. national and international human and veterinary healthcare organizations, complemented by capacity building and a network of disease experts to facilitate risk assessment of detected signals based on the above data.

MOOD outputs are designed and developed together with the end users to ensure their routine use during and after the MOOD project. They are tested and fine-tuned for airborne, vector-borne and waterborne diseases, including antimicrobial resistance. Extensive consultation with end-users, studies on the barriers to data sharing, dissemination and training activities, and studies on the cost-effectiveness of MOOD results will support future sustainable adoption by users.

Services

mundialis has implemented the following aspects of this project to date:

Research and quality analysis of available open geo, meteo and satellite data in Europe
Evaluation and assessment of data sources for the MOOD questions
Improving the spatial resolution of ERA5 data using statistical methods (continental scale)
Provision of Land Surface Temperature (LST), NDVI, EVI satellite data (in particular MODIS)
Extraction of thematic datasets from OpenStreetMap
Preparation of Copernicus-DEM 30m data (as Analysis Ready Data)
Time series analysis of meteorological data (esp. ERA5), incl. Aggregation of data and statistical analysis

Result

Development of a geostatistical method to improve the spatial resolution of ERA5 time series
Development of a method for calculating relative humidity maps from ERA5 time series
Publication of results on the scientific online storage service Zenodo.org
Publication of scientific posters at conferences

Customer

European Commission (EU H2020)

Story

In an environment that is changing due to climate change, animal and human mobility, population growth and urbanization, the risk of new and exotic pathogens going unnoticed is increasing. Traditional infectious disease surveillance methods based on case reporting are often too slow and inaccurate. This is where the EU-H2020 project MOOD comes in. As part of the project, methods and an innovative platform were developed to improve the early detection, assessment and monitoring of infectious diseases in Europe. By using big data from a variety of sources such as geo, satellite, weather and social media data, MOOD enables a more precise and faster response to disease outbreaks.