Medonna- Medication Tracker and Reminder
Link to open source: https://github.com/seemaparveeninsha2004-alt/Medonna
Link to Live Project: https://medonna.vercel.app/
Medonna Mobile App
Many patients today struggle with remembering their medications, managing complex schedules or understanding what to do when they miss a dose. These challenges often lead to poor treatment outcomes and increased health risks. To address this widespread issue, the “ Medonna Mobile App-Medication Reminder & Tracker using Conversational AI” introduces an intelligent voice-driven healthcare assistant that provides personalized, continuous and adaptive support. Unlike traditional reminder apps that simply notify users this system uses real conversation and emotional understanding to guide patients in a compassionate human-like manner.
The assistant is built using Agora Conversational AI enabling natural low-latency voice interactions with speech-to-text, text-to-speech, emotion detection and contextual understanding. The system conducts real-time conversations reminds users of doses, asks for confirmations and responds based on tone and intent. If emotional cues suggest stress or confusion the AI modifies its conversation, offers clarity, or triggers caregiver alerts.
The mobile application, developed in Flutter provides two interaction modes:
-
Easy Mode with a voice-first UI, oversized elements and minimal navigation for elderly or disabled users.
-
Normal Mode with dashboards, analytics, adherence history and caregiver panels.
On the backend, Supabase handles authentication, secure data storage, and medication schedule management through PostgreSQL tables. Supabase Edge Functions process AI intents, route caregiver alerts and handle predictive logic. A notification pipeline using Firebase push notifications with SMS and email fallback ensures reliable delivery even during poor connectivity.
A key innovation is the predictive adherence engine powered by behavioral pattern detection. It analyzes factors such as time-of-day patterns, repeated delays, emotional tone from voice interactions and historical logs to anticipate missed doses. The system then suggests optimized schedules or micro-reminders making adherence proactive instead of reactive.
By combining conversational AI, real-time analytics, caregiver connectivity and a robust technical architecture, this project delivers a compassionate, intelligent and secure solution that significantly enhances medication safety and patient wellbeing.
This build was uploaded as a hackathon project













