UNICEF adopts machine learning to boost Africa vaccination
The United Nations Children's Fund (UNICEF) is implementing machine learning to expedite immunisation programs in Central and West Africa.
This falls under the Reach the Unreached (RtU) pilot initiative, which was launched in Cameroon, Chad, Guinea, and Mali.
The program uses machine learning technologies to disaggregate population data to estimate vaccination coverage.
RtU is collaborating with the Frontier Data Network (FDN).
UNICEF officials explain that using this approach, colleagues in the regional and country offices have mapped over 1.1 million unreached children, with the goal of providing participating countries with an additional, granular source of information to identify local geographies at risk of falling behind, as well as uncovering and investigating child rights inequities, beginning with immunisation and birth registration.
“While the proliferation of granular population estimates and vaccination coverage datasets is beneficial and possibly game-changing, these new sources of information will only make an impact for improving health programming and health outcomes if they're integrated into existing information systems and decision-making processes at the country level,” said Rocco Panciera, UNICEF geospatial health specialist.
Manuel Garcia-Herranz, FDN’s principal researcher, noted that without technology, experts lack insights into how data bias and algorithmic inequalities affect combined population estimation and vaccination coverage models.
“Even for single models, understanding performance across different socioeconomic contexts is challenging,” Garcia-Herranz said.