EUVIMEDEUVIMED entdecken
Crossref

A Data-Driven Framework for Predicting Mosquito-Borne Diseases via Earth Observation Data

Dimitrios Sainidis

Aristotle University of Thessaloniki · 01.01.2023

DOI: 10.26262/heal.auth.ir.348136 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Aristotle University of Thessaloniki, erschienen am 01.01.2023. Verfasst von Dimitrios Sainidis.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Mosquitoes are considered the deadliest animals in the world, killing an estimated 1 million humans every year. There are about 110 trillion mosquitoes on the planet, approximately 16 thousand for every human, and can carry lethal diseases like Malaria, Dengue, West Nile Virus (WNV), and Zika Virus. The rapid growth of human mobility around the globe, combined with the warming climate and the changing environmental landscape, has led to the expansion of mosquitoes and Mosquito-Borne Diseases (MBDs) into new territories. This growing concern prompted governments worldwide to allocate funds to the research and development of Early Warning Systems (EWS) against the spread of MBDs. Up until now, most studies that develop early warning systems for certain MBDs are applied to large spatial scales, usually, counties or districts, and large temporal resolution usually at time leads of one year to six months. This thesis proposes the development of a data-driven framework based on Machine Learning (ML), for the prediction of WNV cases in a fine spatiotemporal scale (time lead: two weeks, spatial scale: LAU1 or municipality), that allows it to be used operationally as an EWS, and help health authorities in their decision-making. The designed framework relies mainly on historical epidemiological data, satellite-derived Earth Observation (EO) data, and census data. It is applied to two different areas of interest namely the Veneto region in Italy and the Central Macedonia region in Greece. To evaluate the usefulness of this approach, a novel evaluation method is proposed based on the operational application of an EWS. The results show that the data have a high predictive value regarding the prediction of WNV cases and that there is a substantial added value in using the framework operationally as an EWS.

Schlagwörter

mosquito-borne diseasesWest Nile virusearly warning systemsEarth observation datasatellite dataspatiotemporal predictionmachine learningcensus dataepidemiological dataVeneto

Zitieren

APA

Sainidis, D. (2023). A Data-Driven Framework for Predicting Mosquito-Borne Diseases via Earth Observation Data. Aristotle University of Thessaloniki. https://doi.org/10.26262/heal.auth.ir.348136

Vancouver

Sainidis D. A Data-Driven Framework for Predicting Mosquito-Borne Diseases via Earth Observation Data. Aristotle University of Thessaloniki. 2023. doi:10.26262/heal.auth.ir.348136

BibTeX
@article{Sainidis2023,
  author = {Sainidis, Dimitrios},
  title = {A Data-Driven Framework for Predicting Mosquito-Borne Diseases via Earth Observation Data},
  journal = {Aristotle University of Thessaloniki},
  year = {2023},
  doi = {10.26262/heal.auth.ir.348136}
}
Kostenlos starten →