With support from Vaccine Confidence Fund, Macro-Eyes is building and testing a novel machine learning system to predict local vaccine hesitancy among diverse communities in South Africa. The project brief outlines the program overview, objectives and methodology guiding our work.
STRIATA BEHAVIOR enables users to predict human behavior in a specific context, seeing patterns by group as well as by individual. It categorizes human and system dynamics into patterns of adherence, demand, traffic flow, stressors and vulnerabilities. This organizes behaviors into phenotypes that drive predictive models and enable organizations to adjust system behavior for optimal compliance.
Macro-Eyes core technology - STRIATA- is a sector-agnostic supply chain optimization product that has been tested and proven in healthcare supply chains in some of the most challenging environments in the world. STRIATA learns from satellite imagery, demographics, and public data, applying state-of-the-art machine learning to shift the line of sight and reorient supply chains to respond to what comes next, rather than what already occurred.
For doctors and nurses and health workers of all kinds, care is about the individual; administrators and government departments concerned with cost advocate for systems that are simpler to manage - with less variation across populations (and limited personalization). The radical potential of AI is that health systems no longer need to choose between personalization and scale.
In 2018, macro-eyes built a machine learning system that predicts vaccine utilization at the facility, district, and regional level. The Macro-Eyes system cuts the number of errors (under-and over-supply) of the legacy supply chain by 96%. The predictive supply chain for vaccines can be reliably scaled across LMICs utilizing publicly available data with a nominal loss in precision (5%) relative to the EIR- reliant model.