ClotSync

Inspiration

When I first read the problem statement about Thalassemia patients struggling to find blood donors, I decided to research further. What I discovered shocked me:

1. India has over 40 million Thalassemia carriers, yet thousands of patients die each year due to blood shortages.
2. Blood donation systems today are reactive, not proactive—patients must beg for donors on social media during emergencies.
3. Nearly 30% of donated blood expires in storage while rural patients travel hours for transfusions. This inefficiency made me realize: We don’t just need another donor-matching app—we need a smarter, predictive system.

What it does

ClotSync is an intelligent blood donation platform that:

1. Uses AI to predict blood shortages before they happen
2. Matches Thalassemia patients with nearby donors in real-time
3. Implements a blockchain-based blood sharing network between hospitals
4. Provides emergency "Blood Now" alerts for critical cases
5. Tracks and rewards recurring donors through a gamified system

How we built it

1. AI/ML: Python with TensorFlow for predictive analytics and demand forecasting
2. Blockchain: Hyperledger Fabric for secure blood inventory tracking between hospitals
3. Mobile/Web: Flutter for cross-platform donor/patient interfaces
4. APIs: Integration with e-RaktKosh and government health databases
5. Security: End-to-end encryption for all sensitive health data

Challenges we ran into

1. Data Limitations: Sparse historical data on blood donation patterns required synthetic data generation
2. System Integration: Overcoming bureaucratic hurdles to connect with government health systems
3. Behavioral Adoption: Designing incentives to change donor behavior from reactive to proactive
4. Privacy Concerns: Balancing data utility with strict patient confidentiality requirements

Accomplishments that we're proud of

1. AI-Powered Prediction Framework – Designed a machine learning model architecture to forecast blood shortages using historical donation patterns and demographic data.
2. Competitive Analysis: 0% existing solutions use real-time hospital inventory sharing. 100% rely on manual donor outreach.
3. Every blood app today is donor-first; we're the first hospital-first system with live inventory swaps to prevent blood expiry (40% waste reduction).

What we learned

1. Predictive models work best when combining donation history with external factors (festivals, weather)
2. Gamification significantly improves donor retention rates
3. Hospitals are willing to share surplus blood when given proper tracking and incentives
4. Simple UI/UX is critical for adoption in rural areas with low digital literacy

What's next for NOW

1. Finalize AI prediction approach
2. Create matching algorithm logic
3. Prepare system architecture
4. Design user flow diagrams
5. Create impact projections

Built With

• aes-256
• azure
• fastapi
• firebase
• google-maps
• hyperledger
• oauth2.0
• python
• pytorch
• scikit-learn
• sqlite
• twilio