Theaim is to build a strong modeling network to strengthen the understanding and ability to predict and model vector-borne diseases (VBD) and future threats in Germany. Focus areas: West Nile fever; tick-borne diseases (Lyme borreliosis, tick-borne encephalitis); diseases transmitted by Aedes mosquitoes (dengue fever, chikungunya).

Goals:

• Involve researchers in VBD modeling through community prediction contests (similar to Kaggle).
• Optimization and integration of data from high-quality sources to facilitate modeling work.
• Mapping and forecasting the spread of VBD in Germany using different data sources.
• Projection of short-term trends in vector distribution and VBD risk in Germany.
• Evaluation of the implementation of various intervention scenarios, including vaccination programs in Germany.

Sub-projects:
SP1: Coordination, stakeholder engagement and capacity building.
SP2: Data integration and optimization
SP3: Early warning and short-term forecasting
SP4: Scenario-based modeling
SP5: Health economic modeling

modeling approach:

1. Process-based models – Dynamic (vector) population models using life cycle data.
2. Statistical models – Generalized additive models (GAM), latent Gaussian models and INLA for spatiotemporal data.
3. Machine learning methods – Extreme Gradient Boosting (XGBoost).
4. Multi-model ensemble – Higher prediction accuracy and robustness than with individual models.