Research Initiative · Zanzibar, Tanzania

Precision tools for
malaria elimination

MalarAI integrates artificial intelligence, human mobility analytics, and geospatial modelling to identify hidden transmission sources, map importation pathways, and accelerate the end of malaria.

Population dynamics High-resolution modelling of human movement and mobility
AutoML Pipeline Continuously learning, near real-time surveillance
Open Source Transparent, reproducible, open source methods

Turning complex data into targeted action

Integrating surveillance, mobility, genomics, climate, and environmental data to pinpoint where transmission actually occurs — not just where cases reside.

Dynamic Population Mapping

High-resolution population and mobility models that reveal how people move — and how malaria parasites move with them — integrating smartphone GPS, Meta mobility data, and WorldPop datasets.

Learn more →

AI-Driven Surveillance

An AutoML pipeline that continuously integrates case surveillance, climate, mobility, and genomic data to identify micro-transmission sources, map importation pathways, and forecast outbreak risk in near real-time.

Learn more →

Resistance Modelling

Spatiotemporal forecasting of drug and insecticide resistance across endemic regions — helping programmes stay ahead of emerging threats before treatment and vector control options are compromised.

Learn more →

Breaking through elimination stagnation

Despite two decades of intensive interventions, malaria persists in Zanzibar. Transmission is concentrated among young men aged 15–45, driven largely by outdoor exposure and importation from mainland Tanzania — dynamics that existing surveillance systems were not designed to detect.

MalarAI is developing AI-driven tools — built in direct partnership with the Zanzibar Malaria Elimination Programme (ZAMEP) — to identify the true sources of transmission and enable precision targeting of interventions where they are needed most.

Read about our approach

Open & reproducible

All methods published with transparent assumptions and explicit uncertainty estimates.

Partnership-driven

Co-designed with ZAMEP and national malaria programmes across the region.

Actionable outputs

Built for real-world planning, targeting, and evaluation — not just publication.

Capacity building

Training local programme staff to own and sustain AI tools beyond the project.

Led by world-class researchers

MalarAI brings together spatial epidemiologists, global health experts, and data scientists from leading institutions worldwide.

University of Southampton — WorldPop
London School of Hygiene & Tropical Medicine
Johns Hopkins University
ZAMEP, Zanzibar
Meet the team

Supported by leading institutions

Web Science Institute, University of Southampton

Pilot funding provided by the Web Science Institute Pilot Project Fund 2025–26.

Interested in collaborating?

We work with national programmes, research partners, and data providers around the world.

Get in touch