The challenge
Frontline clinicians needed a decision-support tool that works before a malaria test is run, including in facilities where connectivity is unreliable. Waiting on the cloud was not an option.
What we built
Malaria Clinical Assistant Prototype is an edge-device malaria prediction app. Clinical intake and screening run on the phone with a locally deployed ML model, while SQLite keeps patient and visit data available offline as a decision-support layer before laboratory confirmation.
- On-device ML inference for malaria risk screening
- SQLite local storage for offline clinical workflows
- Structured intake for facility, demographics, and visit data
- Sync-ready design when connectivity returns
Outcome
Clinicians get a portable decision-support assistant that keeps working without the network, so screening guidance is available at the point of care before testing.