This effort is part of the Reach the Unreached (RTU) pilot program, which has been initiated in Cameroon, Chad, Guinea, and Mali.
The program employs machine learning technologies to analyze population data and provide more precise estimates of vaccination coverage.
RTU collaborates with the Frontier Data Network (FDN). UNICEF representatives note that this strategy has allowed regional and country office teams to identify over 1.1 million children who have not been reached.
The objective is to equip participating nations with comprehensive data to pinpoint areas at risk of lagging behind and to tackle disparities in child rights, beginning with immunization and birth registration.
“While the spread of granular population estimates and vaccination coverage datasets is beneficial and potentially game-changing, their impact on improving health programming and outcomes will only be realised if integrated into existing information systems and decision-making processes at the country level,” stated Rocco Panciera, UNICEF's geospatial health specialist.
Manuel Garcia-Herranz, principal researcher at FDN, stressed that without technological support, experts are unable to fully understand how data biases and algorithmic inequalities influence combined population estimation and vaccination coverage models.
“Even for single models, understanding performance across different socioeconomic contexts is challenging,” Garcia-Herranz added.