Spatio-temporal variation of malaria transmission in Africa: modelling, capacity-building and translating research into practice

Objectives

The two main aims of this proposal are:

1. To develop computationally efficient algorithms for malaria prevalence mapping.
2. To make such methodology available and accessible to researchers and public health staff at the Ministries of Health in Kenya, Somalia, Sudan, Tanzania and Uganda

To this end, the following specific objectives will be pursued:

1. To implement approximations to spatio-temporal Gaussian processes for malaria mapping in sub-Saharan Africa using the TensorFlow computing platform.
2. To validate each of the approximations in 1) using data on malaria prevalence collected in Africa since 1900, which are publicly available from the Harvard dataverse.
3. To develop a user-friendly web application, Maplaria, that will allow Ministry of Health staff to independently perform prevalence mapping.
4. To build a capacity platform between Lancaster and KEMRI Wellcome Trust in Nairobi through training activities and development of joint research projects.