FLUTTER ECOSYSTEM

madhurchouhan7/synovia_ai_telehealth_app

Synoviaは、ユーザーが症状を確認し、パーソナライズされた健康インサイトを得て、リアルタイムでスマートな医療アドバイスを受けられるAI駆動のモバイルヘルスアシスタントです。GoogleのGemini APIを統合することで、アプリは知能的に以下を行うことができます:ユーザーが入力した症状を分析する、可能性のある健康状態を予測する、次のステップの提案を行う。

synovia_ai_telehealth_app のプロジェクト画像
Stars
2
Forks
1
最終プッシュ(UTC)
2025/09/24
プロジェクト状態
公開中
Madhur  GitHub avatar
GITHUB User

Madhur ↗

Building AI/ML systems, LLMs & developer tools. ECE student | Researcher | Open-source learner

言語DartC++CMakeJavaScriptSwiftCHTMLKotlinObjective-C

技術トピック

使用している依存関係

元の README

以下は英語原文のスナップショットです。最新版は GitHub をご覧ください。

README を開く / 閉じる

🧠 Synovia AI — Your Smart Telehealth Companion

Synovia AI is an AI-powered telehealth mobile application built using Flutter, Firebase, and Google's Gemini AI. It helps users assess symptoms, track health conditions, and receive personalized AI-driven health insights—all from the comfort of their smartphones.

🌐 Landing Page(https://synovia-landing-page.vercel.app/)
▶️ Demo video(https://youtube.com/shorts/ee1Y5Ld0jF0?feature=share)

🚀 Features

  • 🔍 AI Symptom Analysis: Get instant responses and insights using Gemini-powered AI chat.
  • 📈 Health Tracker: View and monitor symptom severity trends over time.
  • 🗺️ Nearby Doctor Finder: Access a curated list of doctors near your location (coming soon).
  • 🧠 Personalized Responses: Tailored recommendations based on symptom data and chat history.
  • 💬 Conversational UI: Clean, chat-first design for an intuitive user experience.
  • 🗂️ Secure & Fast: Built using Firebase Auth, Firestore, and serverless backend functions.

📱 Screenshots

Symptom Chat Health Insight AI Response Doctor Finder

https://raw.githubusercontent.com/madhurchouhan7/synovia_ai_telehealth_app/HEAD/assets/screenshots/chat_page.png https://raw.githubusercontent.com/madhurchouhan7/synovia_ai_telehealth_app/HEAD/assets/screenshots/chat_page2.png https://raw.githubusercontent.com/madhurchouhan7/synovia_ai_telehealth_app/HEAD/assets/screenshots/health_reports.png

https://raw.githubusercontent.com/madhurchouhan7/synovia_ai_telehealth_app/HEAD/assets/screenshots/home_page.png https://raw.githubusercontent.com/madhurchouhan7/synovia_ai_telehealth_app/HEAD/assets/screenshots/nearby_doctors.png https://raw.githubusercontent.com/madhurchouhan7/synovia_ai_telehealth_app/HEAD/assets/screenshots/nearby_doctors2.png https://raw.githubusercontent.com/madhurchouhan7/synovia_ai_telehealth_app/HEAD/assets/screenshots/health_progress.png


🧰 Tech Stack

  • Frontend: Flutter (Dart)
  • Backend: Firebase Functions (Node.js), Firestore, Firebase Auth
  • AI Model: Gemini Pro (Google AI)
  • State Management: Provider
  • Storage: Hive (Local), Firebase Storage (Cloud)
  • Maps & Location: Google Maps API
  • Analytics & Crash Reporting: Firebase Analytics + Crashlytics

📂 Project Structure

lib/ ├── main.dart ├── config/ ├── core/ ├── features/ ├── utils/ functions/ # Firebase Cloud Functions (Node.js) assets/ # Images, icons, illustrations


💡 How it Works

  1. User signs in securely via Firebase Auth.
  2. User inputs symptoms through chat UI.
  3. Gemini AI model analyzes symptoms and generates insights.
  4. Responses are stored locally in Hive and backed up in Firestore.
  5. Health severity trends are visualized.
  6. (Upcoming) Doctor list is fetched based on location & condition.

🚧 Roadmap

  • [x] MVP with AI chatbot & health tracker
  • [x] Firestore + Hive integration
  • [x] Firebase Auth & onboarding
  • [x] Nearby doctor suggestions (location-based)
  • [x] Disease precautions integration
  • [x] Landing Page with waitlist + app demo
  • [ ] Launch on Google Play Store
  • [ ] iOS build & deployment

🛠️ Run Locally

git clone https://github.com/madhurchouhan7/synovia_ai_telehealth_app.git
cd synovia_ai_telehealth_app
flutter pub get
flutter run

🌐 Live Demo

  • 🚀 Coming soon on Google Play Store
  • 📩 Sign up for early access: synovia.health (Launching soon!)

🙋‍♂️ Author

  • Built with ❤️ by Madhur Chouhan
  • 📬 Reach out: madhurchouhan7@gmail.com

🤝 Contributing

Pull requests are welcome! For major changes, please open an issue first. Let's make digital healthcare smarter — together!