What it does

Dengue Radar helps control officers to decide where to act, what to do and where the virus will go next.

Example of how it works:

-3D City Risk Predictor & Resource Allocator: Officer enters number of teams available (e.g. “I have 2 teams”). The model ranks all zones in terms of risk, taking into account the reproduction rate (R₀) and the number of stagnant water basins, and chooses the best targets for that capacity. In our demo, clearing Area B (92% risk, R₀ 2.8) and Area D (78% risk, biggest larval hotspot) reduces district-wide transmission by 74%.

-Dengue Memory: It compares today's humidity, rainfall, and case trajectory to past outbreaks and mimics what worked. In our demo, Area B represents the July 2019 outbreak at 92.4%. Street fogging failed then but larvicide in indoor drains arrested 70% of cases. The system then tells the officer to skip the fogging and apply the larvicide within 48 hours.

-Hidden Connections Mosquitoes have a range of 200m, but people can travel by train for 15km. Using mobility data it finds that 14,200 people commute daily from outbreak-hit Area B (84 cases) to Area A. Area A has only 4 cases and appears safe on a normal heatmap, but the system predicts a wave there within 7 days. It then recommends spraying of gutters within 300m of the Station A exits and to send SMS alerts to commuters on that route.

Built With

Share this project:

Updates

Submission history