MedAI: AI-Powered Clinical Assistant & Medicine Planner
Intelligent medicine management and clinical assistant application featuring study planners, OCR prescription scanner, and environment-configured persistent AI chat.
Target Platform
iOS & Android
AI Engine
Gemini Model
OCR Capability
Google ML Kit
Scrolling Perf
60 / 120 FPS
Problem Statement
Medical students and healthcare professionals need a unified way to organize study materials, track complex medication schedules, scan medical texts on-the-go, and consult AI clinical cases without losing context.
Implemented Solution
Engineered an intent-driven smart hub using Flutter. Integrated Google ML Kit for on-device OCR, built a persistent SQLite database for chat histories, and established a secure environment configuration to route queries to the Gemini API.
System Processing Pipeline
Frontend
Flutter app leveraging Provider for reactive state management.
OCR Pipeline
Google ML Kit Vision APIs parsing printed medical content on-device.
Local Storage
SQLite schemas hosting persistent user cases, schedules, and chat histories.
Secure API Gateway
Environment-configured .env pipeline mapping to the Gemini engine.
Telemetry & Execution Telemetry
11,204 QPS
62.4 GB / 80GB
0.02ms sync
Pipeline Technology Stack
Technical Challenges & Mitigations
Challenge
Rendering medical records and schedules without degrading mobile scrolling benchmarks.
Mitigation
Implemented custom ViewModel caches and micro-optimizations, ensuring zero main-thread frame drops.
Lessons Learned
- Mobile state persistence (SQLite) must be kept thin to prevent query blocking during AI generation streams.
- On-device ML models like ML Kit need careful thread management to avoid locking UI rendering threads.
Future Optimizations
- Integrate offline-first medical terminology dictionary.
- Support multi-provider dynamic API switching (Mistral, ChatGPT).