Inspiration
Seeing how delayed outbreak responses cost lives in Nigeria inspired me to build a tool that gives health workers and communities early warning they can trust.
What it does
Outbreak sentinel predicts disease outbreak weeks in advance, models how they might spread, identifies high-risk LGA, and sends real-time alerts to health Workers and communities even in low connective areas.
How we built it
I used hybrid AI models (time-series and spatial analysis), integrated multiple data sources (NCDC, satellite, mobile), and created simple mobile/SMS alerts for quick, actionable information
Challenges we ran into
I tracked fragmented data, unreliable connectivity, and making sure the system stays affordable and compliant with Nigeria's data protection laws.
Accomplishments that we're proud of
What we learned
I learned how to combine AI, satellite data, and local reports into a practical, low -bandwidth solution that works in areas with limited infrastructure.
What's next for outbreak sentinel
Pilot in high-risk states add vaccine tracking, expand to west Africa and keep improving accuracy with community feedback.
Built With
- amazon-web-services
- flask
- javascript
- mongodb
- ncdc
- postgresql
- python
- react
- sormas
- telco
- tensorflow
- twilio