Surveilling Malaria through machine learning and clustering tools in pregnancy

The vision of the World Health Organisation (WHO) for 2030 is a world free of malaria.

For this, agile and robust malaria surveillance systems are required to efficiently guide actions towards interruption of transmission.

Estimating malaria trends from passive detection of clinical malaria cases at health facilities or from cross-sectional surveys remains difficult and expensive.

Pregnant women represent a promising convenience group for malaria surveillance, providing a representative section of the overall population in a cost-efficient and sustainable manner.

Serological and molecular surveillance has also become a potential key approach to guide elimination efforts, providing information about the history of exposure, the geographic origin (malaria importation) and the intensity of malaria transmission.

Here we propose to develop and apply novel statistical tools (adapted from the field of cosmology) to test an innovative and cost-efficient surveillance approach based on the strategic use of parasitological, serological and genomic data from easy-access pregnant women at antenatal care (ANC) clinics.

The application of these new tools on data obtained from pregnant women can suppose an enormous breakthrough for sustainable and actionable surveillance systems that can accelerate efforts towards malaria elimination.

With these new developed tools I will a) assess the potential of parasitological and serological data from pregnant women at first ANC visit as a source of reliable data to reflect temporal and spatial malaria trends in the community and b) compare genetic metrics in the parasite population of pregnant women and the overall community that can inform about changes of malaria transmission, clustering of infections and parasite importation.