Project Overview
MediSure AI is an on-device AI-powered medicine authentication app built for AfroEdge Africa. In Africa, 1 in 10 medical products is substandard or falsified — putting millions of lives at risk. MediSure uses advanced AI and military-grade security to instantly verify any medication's authenticity through a simple QR or serial number scan, protecting users and their families from counterfeit drugs — with zero internet required.
The Engineering Challenge
- 1 in 10 medical products in Africa is substandard or falsified, creating a life-threatening public health crisis with no accessible, affordable verification tool for ordinary consumers.
- Internet infrastructure is unreliable across Africa — a drug verification app that requires connectivity would fail precisely in the remote areas where counterfeit drugs are most prevalent.
- On-device AI inference needed to be fast enough for real-world use (millisecond-level) on low-end Android devices with limited RAM and processing power.
- Medical data handling demanded military-grade security with zero raw data transmission — AES-256-GCM encryption, SHA-256 hashing, certificate pinning, and digital signatures were non-negotiable.
- The AI needed to continuously improve its counterfeit detection accuracy without compromising user privacy — requiring federated learning rather than centralised data collection.
Our Solution
- Built on-device TensorFlow Lite AI engine for instant product authentication — processes QR scans and serial numbers in milliseconds with advanced fraud pattern detection and real-time risk scoring, fully offline.
- Implemented QR Code & Serial Number Verification with instant authenticity status, detailed product information, and anomaly flagging including duplicate serial number detection and geographic pattern analysis.
- Applied AES-256-GCM encryption with SHA-256 hashing, certificate pinning, and digital signatures across all data storage and sync operations — zero raw data ever leaves the device.
- Built Self-Learning AI with federated learning — continuously improves counterfeit detection by analysing confirmed cases from regulatory bodies across Africa via secure OTA model updates.
- Designed privacy-by-default architecture: no PII stored or synchronised at any stage, encrypted local database for offline verification history, seamless background sync when connectivity is restored.
- Automated CI/CD deployment pipeline for Android, iOS, and backend — with real-time error tracking, crash reporting, and performance benchmarking (AI speed, audio latency, battery consumption).
Results & Outcomes
- Full offline drug verification using on-device TFLite AI — no internet required
- Military-grade AES-256-GCM encryption with zero raw data transmission
- Federated learning improving counterfeit detection accuracy without touching PII
- QR code and serial number scanning with millisecond-level response time
- Privacy-by-design architecture — no personal data stored or synchronised
- Deployed live on iOS (App Store) and Android (Google Play)
App Screenshots
Why AfroEdge Africa Chose FoogleTech Software
FoogleTech Software is a specialist engineering company with over a decade of expertise in AI, embedded systems, IoT, and full-stack software development — serving product teams and enterprises across AI & Machine Learning and beyond. Our engineers don't just write code — they understand the domain, the constraints, and the real-world pressures that ship deadlines create. For AfroEdge Africa, that meant deploying a pre-vetted team with direct experience in Mobile App, TensorFlow Lite, Healthcare AI, reducing ramp-up time from months to days and delivering production-quality work from the first sprint.
Every FoogleTech engagement starts with a structured discovery phase, follows a disciplined agile delivery model with daily engineering syncs, and ends with complete documentation handover — so your in-house team owns the outcome. No black boxes, no lock-in.