High-Granularity, Machine Learning Informed Spatial Predictive Model for Epidemic Monitoring

Strategic Insights -

Research from Polytechnic University Milan has developed a predictive model for real-time monitoring of epidemic dynamics at the municipal scale in Lombardy region, Italy. The model leverages Emergency Medical Services (EMS) dispatch data and Geographic Information Systems (GIS) methodologies to categorize municipalities into five epidemic scenarios. The system is designed for operational applicability, emphasizing simplicity, speed, and interpretability.

Key Takeaways:

  • The model uses EMS call data as an early proxy for outbreak detection, offering a high-granularity approach that is not limited by institutional delays.
  • The system integrates spatial filtering and machine learning (random forest classifier) to categorize municipalities into five epidemic scenarios: no diffusion, active spread, and increasing trends.
  • The model was developed in collaboration with the Lombardy EMS agency (AREU) and is designed for operational applicability.
  • The research found that the model shows promising predictive capacity, particularly for identifying outbreak-free areas.
  • Performance of the model is affected by changing epidemic dynamics, such as those induced by widespread vaccination.
  • The framework supports health decision-makers with timely, localized insights, offering a scalable tool for epidemic preparedness and response.

Statistics:

  • Five epidemic scenarios are categorized by the model: no diffusion, active spread, increasing trends, and widespread spread.
  • The model uses EMS call data to detect outbreaks, with a reported accuracy of 87.2%.
  • The model was developed in collaboration with the Lombardy EMS agency (AREU), which collected and analyzed EMS dispatch data for the study.
  • The study found that the model performed best in identifying outbreak-free areas, with a reported accuracy of 95.1%.
  • The model was tested on data from the Lombardy region, Italy, with a population of approximately 10 million people.

Sources:

  • A High-Granularity, Machine Learning Informed Spatial Predictive Model for Epidemic Monitoring: The Case of COVID-19 in Lombardy Region, Italy. Applied Sciences, 2025, 15(15):8729. (Applied Sciences - http://www.mdpi.com/journal/applsci).
  • Lorenzo Gianquintieri, et al. "A High-Granularity, Machine Learning Informed Spatial Predictive Model for Epidemic Monitoring: The Case of COVID-19 in Lombardy Region, Italy." Applied Sciences 2025, 15(15):8729.
  • NewsRx. New Findings from Polytechnic University Milan in the Area of COVID-19 Described. Medical Letter on the CDC & FDA. August 31, 2025; p 378.