Machine learning for blue line tracing and wastewater surveillance
Project Team: Andy Tatem, Somnath Chaudhuri, Shengjie Lai, Kathryn Baxter, Alex Cunningham
Dates: April 2025 – December 2026
Partners: Lead institution: University of Virginia (Mami Taniuchi); partner institutions: Imperial College London (Isobel Blake), Novel-T (Lukas von Tobel), icddr,b (Rashidul Haque), WHO GIS Centre.
WorldPop is partnering with leading international research teams to develop scalable environmental surveillance (ES) approaches for monitoring infectious disease transmission in urban settings. Building on extensive multi-pathogen surveillance data collected in Dhaka, Bangladesh between 2019-2024, this innovative project aims to determine the minimum data requirements for implementing effective ES systems in resource-limited settings.
What is the project trying to accomplish:
Comprehensive sewage surveillance systems, like the one established in Dhaka, have proven effective for monitoring epidemics but require extensive resources for detailed population enumeration and complete sewage network mapping. This approach is logistically and financially challenging to scale across multiple cities, particularly in Africa and other resource-constrained regions.
The project will have a phased methodology approach to develop and validate cost-effective alternatives using AI/Machine Learning-assisted sewage network mapping through extracting features from satellite and drone imagery and determining sewage network details required for effective pathogen surveillance.
WorldPop will lead the exploration of methods for small area population modelling and dynamics in urban areas, using Dhaka’s detailed ground enumeration data undertaken in recent years for the ES work as a gold standard. We will aim to identify optimal microcensus sampling approaches and geospatial modelling frameworks that can give us the best accuracy with limited resource demands for ground enumeration.
Following successful validation in Dhaka, Phase 2 (subject to future funding) will implement these optimised approaches in selected African cities, potentially expanding to cities in Nigeria, the Democratic Republic of Congo and South Africa where ES is being trialled or implemented.
Link image
Sanitation and hygiene in Rohingya camp by Rabiul Hasan/icddr,b – CC BY-NC-ND 2.0
About Us
The WorldPop research programme, based in the School of Geography and Environmental Sciences at the University of Southampton, is a multi-sectoral team of researchers, technicians and project specialists that produces data on population distributions and characteristics at high spatial resolution.
Initiated in October 2013 to combine The AfriPop Project, AsiaPop and AmeriPop projects, we have a diverse portfolio of projects, including large multi-million-pound collaborative projects with partner organisations, commercial data providers and international development organisations.