Amref Health Africa predicts and prevents malnutrition hotspots in Kenya
Amref Health Africa, a global health nonprofit, implemented an Azure AI-powered malnutrition forecasting tool in partnership with the Kenyan Ministry of Health, the University of Southern California, and Microsoft AI for Good Lab. The tool ingests large volumes of anonymized health data from Kenya's District Health Information System (DHIS2), satellite imagery, and other public datasets. AI models trained on historical health and environmental data forecast malnutrition severity and hotspots at sub-county level for 1, 3, and 6 months ahead. This enables Amref and partners to plan interventions, mobilize resources, and pre-position supplies ahead of droughts, floods, or food insecurity events. Healthcare workers gain critical, location-specific insights for early action, aiming to reduce malnutrition and its long-term effects in children and vulnerable groups. The collaborative effort demonstrates Microsoft Azure AI's power in transforming public health with predictive and preventive care. Amref plans to expand and replicate the AI solution to other countries and humanitarian crises where health data is available.
- Organization
- Amref Health Africa
- Industry
- Healthcare
- Location
- Kenya
- Published
- December 2025
Reported outcomes
Strategic outcomes
Why do we believe this deployment?Customer identity, provider attribution, maturity, and source checks
- Customer
- Amref Health Africa, Kenya Ministry of Health
- Provider
- Microsoft
- Maturity
- Unknown
- Linked source
- Microsoft Customer Stories
No explicit deployment-stage evidence found.
Primary read
Use case focus
Showing 3 of 3
- 1Malnutrition forecasting tool for public health planning
- 2AI-driven preventative resource allocation in healthcare
- 3Early warning system for child malnutrition hotspots
- High rates of malnutrition, especially among children under five, with 18% stunted (2M+) due to chronic malnutrition.
- Difficulty in monitoring and predicting future hotspots due to data silos and manual data ingestion.
- Limited ability for proactive intervention and resource allocation during droughts and floods.
- Healthcare workers lacked timely insights to deploy nutritional and medical resources strategically.
- Large volumes of health and environmental data were underutilized in planning.
- Developed AI models using Microsoft Azure AI and satellite imagery, plus Kenya Ministry of Health clinical data (DHIS2).
- Tool predicts malnutrition risk by location and time interval (1, 3, and 6 months), using IPC-AMN scale.
- Partners (Amref, Kenyan Ministry of Health, USC, Microsoft AI for Good Lab) collaborate on data-driven, predictive solution.
- Secure data storage and model training in Azure Cloud.
- Share predictions and forecasts with community health workers and other nonprofits for targeted interventions.
- Earlier intervention and pre-positioning of supplies in high-risk areas.
- Improved planning and resource allocation for healthcare organizations.
- Shift from reactive responses to predictive, preventative actions.
- Potential nationwide reduction in malnutrition rates, improved child health outcomes.
- Monthly risk analysis, compared to twice-yearly manual reviews.
Architecture
Health and environmental data from DHIS2 and satellite imagery are aggregated and anonymized. Models trained and tested in Microsoft Azure AI and AI for Good Lab forecast malnutrition risk (IPC-AMN scale) for specific sub-counties and timeframes. Predictions are provided to Amref, government bodies, and nonprofit partners, enabling targeted interventions.
Implementation partners2
Sources & evidence1
- Customer explicitly identified
- Primary source available
- Technical implementation details available
AI-generated summary. Verify important details with the linked sources before relying on this case.
Explore related AI use cases
Was this useful?
Community
Comments
No published comments yet.