Inspiration We were inspired by the increasing need for real-time healthcare solutions for the elderly, chronically ill, and vulnerable individuals who live alone or lack immediate access to care. Existing solutions often depend on wearables, which are costly or inconvenient. We envisioned a smart, AI-powered mobile app that requires nothing more than a smartphone to monitor health, detect emergencies, and connect users to help instantly.
🚀 What it does SMART HEALTH GUARD is a Flutter-based AI-powered healthcare app that provides:
Heart rate monitoring using the phone’s camera and flashlight via PPG (no external devices needed)
Emergency detection like fall alerts using a CNN-based ML model
Auto-alerts to emergency contacts and services, including live location sharing and alarms
Medication reminders with notifications to keep users on track
Smart home control for fans, lights, and ACs — ensuring comfort and safety
Voice-enabled commands and alerts for elderly-friendly accessibility
🏗️ How we built it Flutter for cross-platform mobile development
Camera + Flashlight to capture PPG signals for heart rate measurement
TensorFlow Lite CNN model for fall detection
Syncfusion Charts to display health trends
Flutter TTS/STT for voice commands and accessibility
HTTP + MQTT for smart appliance communication
Geolocator + Local Notifications for emergency response and reminders
Shared Preferences for storing user data locally
😅 Challenges we ran into Accurately detecting heart rate through PPG using smartphone camera and flashlight
Training and integrating a lightweight yet reliable fall detection ML model
Balancing real-time performance with low power consumption
Ensuring the UI was intuitive and accessible for elderly users
Managing permission access for sensitive features (camera, mic, location) securely
🏅 Accomplishments that we're proud of Successfully implemented real-time heart rate monitoring without any external devices
Built a working fall detection system with emergency alerts and geolocation sharing
Designed a clean, accessible UI for users of all age groups
Integrated smart appliance controls with MQTT for IoT interaction
Achieved a modular architecture allowing easy future enhancements
📚 What we learned Learned how to leverage smartphone hardware (camera, flashlight) for health-tech use cases
Gained experience with real-time ML model integration in Flutter using TensorFlow Lite
Understood the complexities of designing for elderly users, including accessibility and simplicity
Discovered best practices for offline-first health applications
Improved our ability to collaborate on a multi-functional app within a time-constrained hackathon
🔮 What's next for Smart Health Guard Enhance emergency detection accuracy using sensor fusion with wearables
Integrate with platforms like Google Fit and Apple Health for data syncing
Develop offline voice command recognition for full accessibility
Add predictive health alerts using historical data and AI models
Launch on Google Play Store with multi-language support and caregiver dashboards
Log in or sign up for Devpost to join the conversation.