HemoLink

"HemoLink – Where Data Meets Humanity"

Comment

Inspiration

Project Story: HemoLink – AI-Powered Companion for Thalassemia Warriors

About the Project

HemoLink is an AI-powered mobile application designed to support Thalassemia patients and amplify the efforts of Blood Warriors, a dedicated volunteer network. The app bridges critical gaps between **patients, donors, hospitals, and support systemsby offering features like:

  • Real-time blood donor connectivity
  • AI-based donor availability prediction
  • Health education in regional languages
  • Mental and emotional support tools

All this is achieved while maintaining data security, **accessibility, and **inclusivity for diverse user groups.

What Inspired Us

Our inspiration stemmed from real-life conversations with Thalassemia-affected families during a blood donation camp. One particularly moving story involved a mother who traveled over 100 km every few weeks, uncertain if blood would be available for her child.

This led us to ask:

“Can we use AI and real-time technology to reduce the uncertainty and improve lives for such families?”

That question sparked the creation of HemoLink — a digital solution to support lives, one link at a time.

What We Learned

This project taught us valuable technical and human lessons:

  • Empathy in design is essential — direct interviews guided feature development.
  • AI can save lives, even with simple predictive models.
  • Privacy and encryption must be prioritized when handling health data.
  • Collaboration with NGOs and doctors is crucial for realistic workflows.

How We Built It

We followed a modular approach, prioritizing real-world user needs and scalability.

🔧 Tech Stack

Component Technology Used
Frontend Flutter (Android/iOS)
Backend Node.js + Firebase
Database Firestore (NoSQL)
AI Models Python (scikit-learn, TF)
Authentication Firebase Auth
Notifications Firebase Cloud Messaging
Hosting Google Cloud
APIs Integrated e-RaktKosh (mocked), Blood Bridge

AI Logic

We implemented a basic AI model to predict donor re-availability based on historical data.

Let:

  • ( f ) = past donation frequency
  • ( d ) = days since last donation
  • ( l ) = location proximity
  • ( h ) = donor's health score

Then the probability ( P ) of a donor being available is modeled as:

$$ P(\text{available}) = \sigma(w_1 f + w_2 d + w_3 l + w_4 h + b) $$

Where ( \sigma ) is the sigmoid function and ( w_i ) are model weights.

Challenges We Faced

Challenge Solution
Limited public health data Created anonymized sample datasets
Security concerns Used AES encryption & Firebase Auth
Rural accessibility Added offline access + SMS fallback
Multi-language needs Added Hindi, and planned regional expansion
Donor retention Designed gamified rewards system

What Makes HemoLink Special

  • Predictive AI to forecast donor availability and transfusion needs
  • Real-time donor–patient–hospital connection
  • Health education modules in local languages
  • Mental health and peer support features
  • Gamified donor experience to encourage regular donations
  • Offline mode & SMS fallback for rural regions
  • Fully encrypted health records with secure access

Conclusion

Building HemoLink wasn’t just a technical challenge — it was a mission to empower Thalassemia patients and families across India. Through AI, empathy, and design, we developed a solution that has the potential to scale nationally and save thousands of lives.

"Where technology meets empathy, real change begins."

Built With

Share this project:

Updates