DVMA - Dementia Virtual Memory Assistant

Dementia Virtual Memory Assistant

A compassionate mobile app empowering dementia patients, caregivers, and admins

✨ Features • 📸 Screenshots • 🗄 Database • 🔌 API • 🚀 Getting Started • 🌐 Demo

---

📖 About

DVMA (Dementia Virtual Memory Assistant) is an innovative mobile application crafted to support individuals with dementia in navigating daily life, cherishing memories, and fostering connections with loved ones and caregivers. With intuitive role-based access for Patients, Caretakers, and Admins, DVMA delivers real-time notifications, AI-driven chat support, and advanced face recognition to uplift the well-being of dementia patients and their support networks.

---

✨ Features

Category	Highlights
👤 Patient Features	- 📅 Task Management: Create, edit, and track daily/recurring tasks with reminders - 🤖 AI Chat Assistant: Gemini-powered prompts for memory and tasks - 📸 Memory Album: Add photos and notes to relive moments - 👨‍👩‍👧‍👦 Family Management: Add contacts and set emergency numbers - 📍 Location Sharing: Secure sharing with caretakers - 📔 Personal Diary: Private daily reflections with auto-save - 🔔 Smart Notifications: Alerts for tasks and connections - 🎭 Face Recognition: AI-based family identification
🧑‍⚕️ Caretaker Features	- 🔗 Connection Management: Handle patient requests and links - 👤 Patient Monitoring: View tasks, location, and albums - 📞 Direct Communication: Calls and messages to patients - 📸 Advanced Scanner: On-the-go family recognition - 📋 Task Oversight: Remote supervision of tasks - 📊 Reporting System: Submit well-being or feedback reports - 📍 Live Tracking: Real-time patient location
👨‍💼 Admin Features	- 👥 User Management: Edit, ban/unban users and caretakers - 🔔 Global Notifications: Broadcast or targeted alerts - 📈 Reports Dashboard: Review and resolve reports - ⚙️ System Settings: Customize APIs and support - 👤 Account Control: Manage admin permissions - 📊 Analytics: Track engagement and usage patterns

---

📸 Screenshots

Patient Interface

Caretaker Interface

Admin Interface

---

🗄 Database Schema

The Firestore structure uses normalized singular collection names for scalability and real-time sync.

📊 Collections Overview

Collection	Purpose	Key Features
user	Patient profiles and data	Tasks, diary, family, location, notifications
caretaker	Caretaker profiles	Credentials, approvals, connections, notifications
admin	Admin accounts	System access control
connections	User–caretaker links	Status tracking, request/confirm actors, timestamps
reports	User-submitted reports	Issue tracking, moderation workflow
api	System configuration	URLs, support contacts

👤 User Collection

Purpose: Stores patient profiles and related data.

Example Document:

{
  "uid": "abc123def",
  "fullName": "John Doe",
  "username": "johndoe2025",
  "email": "john@example.com",
  "phoneNo": "+1234567890",
  "dob": "1950-05-15",
  "gender": "male",
  "bio": "Loves gardening and music",
  "locality": "Springfield",
  "city": "Los Angeles",
  "state": "California",
  "profileImageUrl": "https://example.com/profile.jpg",
  "isConnected": true,
  "currentConnectionId": "conn789",
  "emergencyContacts": [
    { "name": "Jane Doe", "relation": "Spouse", "number": "+1234567891" }
  ],
  "playerIds": ["oneSignalId123"],
  "isBanned": false
}

Subcollections:

• to_dos: { "task": "Take medicine", "completed": false, "dueDate": "2025-11-02T10:00:00Z" }
• recurring_tasks: { "task": "Morning walk", "dailyDueTime": {"hour": 8, "min": 0} }
• family_members: { "name": "Mary Doe", "relation": "Daughter", "phone": "+1234567892", "imageUrl": "https://..." }
• album: { "title": "Birthday 2025", "imageUrl": "https://example.com/photo.jpg" }
• diary: (Doc ID = date) { "content": "Had a good day today...", "createdAt": "2025-11-02T20:00:00Z" }
• notifications: { "type": "connection_request", "message": "New request", "isRead": false, "createdAt": "2025-11-02T10:00:00Z" }

🧑‍⚕️ Caretaker Collection

Purpose: Manages caretaker profiles and credentials.

Example Document:

{
  "uid": "xyz789abc",
  "fullName": "Sarah Smith",
  "username": "sarahsmith",
  "email": "sarah@example.com",
  "phoneNo": "+0987654321",
  "profileImageUrl": "https://example.com/caretaker.jpg",
  "caregiverType": "nurse",
  "relation": "",
  "experienceYears": 5,
  "experienceBio": "5 years of elderly care experience",
  "graduationOnNursing": "BSN from UCLA",
  "graduationCertificateUrl": "https://example.com/cert.pdf",
  "isApprove": true,
  "isConnected": true,
  "currentConnectionId": "conn789",
  "playerIds": ["oneSignalId456"],
  "isBanned": false
}

Subcollections:

• notifications: { "type": "unbind_request", "message": "Patient wants to unbind", "isRead": false, "createdAt": "2025-11-02T12:00:00Z" }

🔗 Connections Collection

Field	Type	Description	Example
user_uid	String	Patient UID	"abc123def"
caretaker_uid	String	Caretaker UID	"xyz789abc"
status	String	Connection status	"accepted"
timestamp	Timestamp	Request time	"2025-11-02T10:00:00Z"
confirmedBy	String	Who confirmed	"xyz789abc"
requestedBy	String	Who requested	"abc123def"

📊 Reports Collection

Field	Type	Description	Example
sender_uid	String	Reporter UID	"abc123def"
sender_role	String	Reporter role	"user"
reported_uid	String	Reported UID	"xyz789abc"
reported_role	String	Reported role	"caretaker"
title	String	Report title	"Inappropriate behavior during visit"
description	String	Report details	"The caretaker was unresponsive..."
status	String	Review status	"pending"
createdAt	Timestamp	Submission time	"2025-11-02T14:00:00Z"

⚙️ API Collection

Example Document:

{
  "geminiApiUrl": "https://generativelanguage.googleapis.com/v1beta/models/gemini-pro:generateContent",
  "supportEmail": "support@dvma.app",
  "supportPhone": "+1-800-DVMA-HELP",
  "version": "1.0.0",
  "lastUpdated": "2025-11-02T00:00:00Z"
}

---

🔌 API Documentation

DVMA uses Firebase for auth and data, Gemini for AI, and OneSignal for notifications. Endpoints are loaded dynamically from the api collection.

Authentication

• Sign Up/Login: Firebase Auth (createUserWithEmailAndPassword / signInWithEmailAndPassword)
• Role Assignment: Post-signup role selection and profile completion.

Core APIs (Firestore)

• User Tasks: user/{uid}/to_dos (CRUD via SDK)
• Connections: connections (add/update for requests/approvals)
• Notifications: OneSignal push + subcollection storage
• Face Recognition: Client-side ML Kit or Cloud Vision API

AI Integration

• Chat Assistant: POST to Gemini with prompt: "As a compassionate assistant for dementia patients, respond to: {user_input}"

For details: Firebase Docs | Gemini API

---

🚀 Getting Started

Prerequisites

• Flutter SDK (v3.10+)
• Firebase project
• OneSignal account
• Gemini API key

Installation

1. Clone the Repo:

   git clone https://github.com/ajmal-uk/dementia-virtual-memory.git
   cd dementia-virtual-memory

2. Install Dependencies:

   flutter pub get

3. Configure Firebase:

   • Add google-services.json (Android) / GoogleService-Info.plist (iOS)

4. Set Environment Vars:

   • Update lib/config/app_config.dart with API keys

5. Run the App:

   flutter run

Development Tips

• Patient Flow: Test tasks and AI in simulator
• Caretaker: Use Firestore emulator for connections
• Admin: Enable elevated privileges for bans/notifications

Contributions welcome—prioritize accessibility and privacy!

---

🌐 Demo

Download the APK for a hands-on experience. (Replace with actual link for production.)

---

Made with ❤️ for dementia care | © 2025 DVMA Team | Contact