CardiaFit

Personalized Cardiac Rehabilitation: AI-Powered Exercise Plans with Real-Time Pose Correction using Google ML Kit.

Comment

CardiaFit

💡 Inspiration

Cardiac rehabilitation is crucial but often inaccessible, especially for patients in remote areas or those recovering at home. Many lack access to physiotherapists or personalized care. I was inspired to build CardiaFit to bridge this gap using mobile technology to deliver personalized exercise plans and real-time pose correction, ensuring safety and effectiveness during rehab.

🎯 What it does

CardiaFit is a comprehensive mobile application that delivers:

  1. Risk Assessment - Evaluates patient risk based on age, medical history, and physical condition
  2. Customized Exercise Plans - Generates personalized cardiac rehab routines using rule-based scoring systems
  3. Real-time Pose Correction - Uses Google ML Kit to track user movements and correct posture instantly
  4. Complete Guidance System - Provides visual guides, performance feedback, and comprehensive progress tracking

🛠️ How we built it

Frontend Development

  • Developed in Flutter for seamless, accessible user experience across all devices

AI-Powered Pose Detection

  • Integrated Google ML Kit's Pose Detection package to track body joints and form accurately

Intelligent Backend Logic

  • Built Python-based modules for generating risk-adjusted exercise routines

Smart Exercise Generation

  • Implemented rule-based scoring algorithms for personalized workout plans

Professional Animations

  • Created engaging 2D exercise animations using Adobe After Effects and Adobe Illustrator

🚧 Challenges we ran into

  • Cross-Platform Accuracy: Ensuring pose detection worked reliably across different body types, lighting conditions, and mobile devices
  • Real-time Performance: Developing responsive feedback loops without creating lag on budget devices
  • Medical Safety Balance: Designing exercise rules that prioritize safety while maintaining usability
  • Dynamic Personalization: Creating adaptive rehab plans while keeping the interface simple for non-technical users

🏆 Accomplishments that we're proud of

✅ Complete Mobile Solution - Built an end-to-end cardiac rehabilitation system requiring no external hardware
✅ Universal Compatibility - Achieved smooth real-time pose detection even on budget Android devices
✅ Adaptive Intelligence - Created a dynamic exercise engine that adjusts based on user risk and progress
✅ Inclusive Design - Developed accessible UI specifically for elderly users with minimal tech experience
✅ High Accuracy - Achieved excellent pose detection performance across various real-world conditions
✅ User-Friendly Animations - Created intuitive exercise demonstrations that guide without overwhelming
✅ Validated Testing - Successfully conducted initial user testing with non-technical participants

📚 What we learned

  • Healthcare-Tech Integration: How to effectively combine medical logic with ML and app development for real-world impact
  • ML Kit Capabilities: Understanding practical limits and strengths of Google ML Kit in pose tracking applications
  • Medical UI/UX Importance: Learned that simplicity in medical applications can be literally life-saving
  • Innovation Impact: Discovered how small technological innovations can significantly improve patient safety and autonomy

🚀 What's next for CardiaFit

Immediate Enhancements

  • 🎤 Voice-Assisted Guidance - Integrate hands-free voice prompts, especially beneficial for elderly patients
  • ⌚ Wearable Integration - Connect with smartwatches to monitor heart rate, oxygen levels, and fatigue in real-time
  • 👩‍⚕️ Doctor Dashboard - Build web portal for physiotherapists and cardiologists to monitor patient progress remotely

Accessibility & Reach

  • 🌍 Multi-Language Support - Add regional language options for diverse linguistic communities
  • 🎮 Gamification Features - Introduce progress badges, streaks, and motivational messages for consistent engagement
  • 📱 Offline Mode - Optimize for offline access to ensure continuous availability

Advanced Features

  • 🔥 Posture Heatmaps - Visualize pose accuracy over time to identify improvement areas
  • 🚨 Emergency Alert System - Real-time alerts for risky behavior like overexertion or falls detection
  • 🏥 EMR/EHR Integration - Sync rehabilitation data with existing medical records

Clinical & Business Development

  • 🔬 Clinical Trials & Validation - Partner with healthcare institutions for formal medical testing
  • 🤖 AI-Powered Feedback Loop - Use anonymized data to continuously improve pose correction accuracy
  • 💰 Insurance Partnerships - Explore health insurance integration for reimbursement and usage incentives
  • 🩺 Expanded Applications - Adapt platform for stroke rehab, orthopedic recovery, and diabetic exercise management

Built With

Share this project:

Updates