๐Ÿฉธ LifePulse โ€“ Every Pulse Matters, Every Donor Counts

โœ… Inspiration

In India, over 100,000 Thalassemia major patients depend on regular blood transfusions. However, challenges like the unavailability of regular donors, lack of awareness, inadequate medical infrastructure, and fragmented support systems put lives at risk.

We were inspired by the selfless work of organizations like Blood Warriors and decided to build LifePulse, a real-time AI-powered platform that ensures no patient is left waiting for blood.


๐Ÿ’ก What it does

LifePulse is an end-to-end solution that supports Thalassemia patients by:

  • ๐Ÿ”ด Connecting patients to real-time available donors using location and compatibility filters
  • ๐Ÿค– Predicting donor availability using AI models trained on past donation behavior
  • ๐Ÿ‘ฉโ€โš• Facilitating doctor-patient communication via secure chat and teleconsultation
  • ๐Ÿ“š Educating patients and caregivers through regional language content and quizzes
  • ๐Ÿงพ Enabling patients to maintain health records: transfusion dates, medication logs
  • ๐Ÿ” Implementing data privacy protocols with end-to-end encryption

๐Ÿ›  How we built it

๐Ÿ”ง Stack Overview

  • Frontend: React, Tailwind CSS, React Native (Mobile)
  • Backend: Node.js, Express
  • Database: MongoDB
  • Authentication: JWT + OAuth2
  • AI/ML: Python, scikit-learn, Prophet for time-series donor prediction
  • Chatbot: OpenAI API for multilingual support
  • API Integrations:
    • e-RaktKosh for live blood inventory
    • Google Maps API for geo-based donor matching
  • Real-time updates: Firebase Cloud Messaging
  • Encryption: AES-256 for sensitive health and donor data

๐Ÿง— Challenges we ran into

  • Designing a system that ensures low-latency for emergency blood request fulfillment
  • Training an effective donor prediction model with limited historical datasets
  • Integrating government-level systems (e-RaktKosh) without public SDKs
  • Ensuring compliance with healthcare data standards (HIPAA-equivalent)
  • Making the app usable and inclusive for low-tech users in rural areas

๐Ÿ† Accomplishments that we're proud of

  • Built a donor availability predictor with over 80% accuracy
  • Created a multilingual AI chatbot for patient education and help
  • Developed a gamified donor engagement tracker that rewards regular donors
  • Completed a working integration with e-RaktKosh test environment
  • Designed a secure health data infrastructure with access control and encryption

๐Ÿ“š What we learned

  • (\textbf{AI + empathy = social impact}): Technology can only solve human problems when designed for real needs.
  • How to use time-series forecasting for health resource prediction: [ y(t) = g(t) + s(t) + h(t) + \varepsilon_t ] Where:
    • (g(t)): trend function
    • (s(t)): seasonality
    • (h(t)): holidays (e.g., less donation on festivals)
    • (\varepsilon_t): error term
  • The value of clean UI/UX in improving adoption among non-tech-savvy users
  • The challenge of maintaining data security while enabling real-time access

๐Ÿš€ What's next for LifePulse

"Every Pulse Matters, Every Donor Counts."

๐Ÿ“Œ Short-Term

  • Launch MVP with pilot users in collaboration with local NGOs and hospitals
  • Train more accurate AI models using real donor history
  • Add voice-based chatbot for low-literacy areas

๐Ÿ“Œ Long-Term

  • Partner with state health departments
  • Onboard blood banks across India
  • Provide free access for economically weaker Thalassemia patients

Built With

Share this project:

Updates

Submission history