The Disease Incidence and Resource Estimator (DIRE) is a geospatial predictive analytics platform developed to help decision-makers anticipate and prepare for infectious disease outbreaks before they reach a crisis point. By bridging the gap between complex academic forecasting and real-world policy, DIRE provides health officials in Brazil and Peru with a “Mission Control” for managing dengue and malaria.
Rather than simply reporting past data, DIRE uses an ensemble machine-learning approach to provide near-term projections, allowing governments to shift from reactive emergency response to proactive, data-driven preparation.
Tools & Features
Geospatial Predictive
Analytics
Anticipate outbreaks before they peak using a sophisticated machine-learning ensemble. DIRE provides near-term projections (up to two months) of dengue and malaria incidence across multiple administrative levels, from national trends to specific districts.
Note: Information obtained during Feb 2026
Resource & Cost
Estimator
Move from “how many cases” to “how many supplies”. The platform automatically calculates the exact quantity and cost of health resources required, such as vaccine doses, fumigation kits, and medical personnel, based on the predicted case load.
Local Decision Support
PDF reports
Bridge the gap between data and action with downloadable PDF reports. These location-specific summaries translate technical forecasts into clear guidance for local leaders, making it easier to mobilize funding and staff where they are needed most.
Principal Investigator.
"Ultimately, success looks like governments being able to act earlier, moving from reacting to outbreaks to preparing for them."
Why This Matters: The Climate-Health Link
In the Amazon and surrounding regions, the climate crisis is not just an environmental issue, it is a public health emergency. The DIRE platform was built to address the “perfect storm” of factors that are making infectious diseases more frequent and harder to control.
Shifting Habitats & Human Exposure
As climate change and rapid deforestation alter the landscape, they fundamentally change how diseases spread.
Deforestation: Land clearing creates small, stagnant pools of water that serve as ideal breeding grounds for mosquito vectors.
Ecological Shifts: Removing forest cover alters microclimates and mosquito habitats, bringing disease-carrying vectors into closer contact with human settlements.
Migration: As populations move into previously uninhabited areas, human exposure to these shifting transmission zones increases significantly.
Extreme Weather as a Disease Driver
Predicting outbreaks requires understanding the volatile relationship between weather patterns and vector biology.
Flooding & Rainfall: Heavy rains and flooding, often intensified by phenomena such as El Niño, can rapidly expand mosquito breeding sites, leading to sudden surges in case counts.
Temperature Spikes: Warmer temperatures accelerate the life cycle of parasites within mosquitoes, making the insects more infectious quickly and expanding their geographic range.
Protecting the Most Vulnerable
The consequences of these climate-driven surges are not felt equally.
Impact on Children: In 2025 alone, Peru reported 39,000 dengue cases, with a substantial proportion affecting children.
Overwhelmed Systems: Climate-related outbreaks are becoming so frequent that they are overwhelming the current capacity of local governments to respond effectively.
Gender & Health: Pregnant women are at significantly higher risk during these surges, making proactive resource allocation a matter of life and death.
Scientific Rigor & Methodology
To build trust with policymakers and researchers, it is essential to highlight the academic foundation of the DIRE platform. It translates complex climate and health data into actionable insights.
The Ensemble Machine-Learning Model
At the heart of DIRE is an ensemble machine-learning approach first conceptualized by UNICEF and the European Space Agency (ESA). This framework was further developed and validated in a 2024 study published in Scientific Reports.
The model combines multiple predictive algorithms to forecast dengue and malaria incidence rates with high precision. By leveraging an “ensemble” of models rather than just one, the platform provides more stable and reliable projections even when environmental conditions are volatile.
Integrated Data Sources
The platform processes a vast array of data points to generate its two-month outlooks:
