Oct 13 – 15, 2025
Hotel Berlin, Berlin
Europe/Berlin timezone
All review results have been sent out on August 20th + September 2nd.

Leptospirosis Occurrence in Europe: Understanding Environmental and Socio-Economic Drivers Using Machine Learning

Oct 13, 2025, 5:40 PM
15m
Room C4

Room C4

Oral presentation Climate Change & Health Session 4: Climate Change & Health

Speaker

Omid Airom (1. Heidelberg Institute of Global Health (HIGH), Heidelberg University Hospital, Heidelberg University, Heidelberg, Germany. 2. Interdisciplinary Center for Scientific Computing (IWR), Heidelberg University, Heidelberg, Germany.)

Description

Leptospirosis is a rodent-borne infectious disease posing a growing global health threat. Human infections occur through contact with environments contaminated by host urine. Recent ECDC reports indicate a rising trend in Leptospirosis cases across Europe, highlighting the need for improved public health measures. Understanding the factors driving leptospirosis occurrence can enhance surveillance and preparedness actions. Here, we investigate environmental drivers of leptospirosis in Europe using a predictive modelling framework. We implemented an XGBoost model to predict the occurrence of at least one leptospirosis case based on temperature, rainfall, environmental, and socio-economic factors, including population density and GDP. We developed the model using the ECDC leptospirosis patient data at the NUTS3 and monthly resolution. The dataset contains 3,868 occurrence and non-occurrence records from 2009 to 2021. We also conducted a SHAP analysis to understand the feature importance. We achieved 75.97% accuracy, 74.56% precision, and 82.93% recall for the model in prediction, indicating a balanced performance in detecting leptospirosis case occurrences and absences. The SHAP analysis revealed that temperature, livestock population, employment rate, and population density made the most considerable predictive contributions to a case occurrence. Our results contribute to developing spatial risk mapping and prediction to inform healthcare and prevention planning.

Keywords

Leptospirosis, Predictive Modeling, Environmental Drivers, Socio-Economic Factors, Machine Learning

Registration ID 106
Professional Status of the Speaker PhD Student
Junior Scientist Status Yes, I am a Junior Scientist.

Author

Omid Airom (1. Heidelberg Institute of Global Health (HIGH), Heidelberg University Hospital, Heidelberg University, Heidelberg, Germany. 2. Interdisciplinary Center for Scientific Computing (IWR), Heidelberg University, Heidelberg, Germany.)

Co-authors

Prof. Jan C. Samenza (1. Department of Epidemiology and Global Health, Umeå University, Umeå, Sweden. 2. Heidelberg Institute of Global Health, Heidelberg University, Im Neuenheimer Feld 205, Heidelberg, Germany) Dr Prasad Liyanage (Heidelberg Institute of Global Health, Heidelberg University Hospital, Heidelberg, 69120, Germany) Dr Junwen Gou (Department of Epidemiology and Global Health, Umeå University, Umeå, Sweden) Prof. Frauke Ecke (1. Department of Wildlife, Fish, and Environmental Studies, Swedish University of Agricultural Sciences, 901 83 Umeå, Sweden. 2. Organismal and Evolutionary Biology Research Programme, University of Finland, PO Box 65, FIN-00014 University of Helsinki, Finland) Prof. Joacim Rocklöv (1. Heidelberg Institute of Global Health (HIGH), Heidelberg University Hospital, Heidelberg University, Heidelberg, Germany. 2. Interdisciplinary Center for Scientific Computing (IWR), Heidelberg University, Heidelberg, Germany. 3. Department of Epidemiology and Global Health, Umeå University, Umeå, Sweden.) Dr Marina Treskova (1. Heidelberg Institute of Global Health (HIGH), Heidelberg University Hospital, Heidelberg University, Heidelberg, Germany. 2. Interdisciplinary Center for Scientific Computing (IWR), Heidelberg University, Heidelberg, Germany. 3. Department of Epidemiology and Global Health, Umeå University, Umeå, Sweden.)

Presentation materials

There are no materials yet.

Peer reviewing

Paper