I first became deeply concerned about malaria during my work in clinical research and field health programs in East Africa, where I repeatedly saw families in rural and peri-urban communities affected despite ongoing control efforts. What stayed with me was the realization that many infections were not random—they were closely tied to overlooked environmental breeding sites that traditional surveillance often misses.

In the ecosystems where I work, rapid urban expansion, seasonal flooding, and inadequate drainage systems are creating persistent mosquito habitats. At the same time, climate variability is shifting transmission patterns in ways that are difficult to predict using conventional tools alone. Communities often respond reactively, rather than having early, spatially precise warnings of risk.

This project uses AI, drones, and ecological mapping to identify and monitor mosquito breeding sites in real time. Drone imagery will be analyzed using machine learning models to detect stagnant water bodies and high-risk ecological zones. These data will be combined with community reporting and local health surveillance systems to generate targeted intervention maps for vector control teams. Local health workers and community members will be trained to validate and act on these insights.

Within 6–12 months, we expect to produce high-resolution risk maps that guide larval source management and targeted environmental interventions. Over time, this could reduce mosquito density in hotspot areas, lower transmission risk, and strengthen climate-resilient community health systems. Local government teams, community health workers, and environmental officers will be central to implementation, ensuring the system is practical and locally owned.