Inspiration:
Thalassemia patients in India struggle with consistent access to compatible blood, timely medical care, and educational resources. Many families face emotional, logistical, and financial burdens due to fragmented support systems. Blood Warriors is doing incredible work — our idea is to supercharge their impact using AI, data, and smart connectivity to close the gaps in blood donation, awareness, and care.
What it does:
ThalCare AI is a unified, AI-powered platform that: Predicts donor availability using past donation patterns. Instantly matches patients with nearby eligible donors. Allows patients to consult doctors via chat or video. Gamifies donor engagement to boost repeat donation. Educates families with resources tailored for Thalassemia. Integrates with existing systems like e-RaktKosh and Blood Bridge.
How we built it:
We’ve designed ThalCare AI as a modular system:
Frontend: React.js for web and Flutter for mobile. Backend: Node.js with Express + MongoDB. AI Layer: Python-based donor prediction using scikit-learn. Integration: **Google Maps API for geolocation, and simulated integration with e-RaktKosh. **Security: Firebase Auth + JWT with AES encryption for medical data. Live demo setup: Firebase and GCP-hosted backend with sample user flows.
Challenges we ran into:
Lack of open-access real-time donor data made training the AI models harder — we simulated datasets. Mapping government systems (like e-RaktKosh) without access to full APIs. Balancing simplicity for rural users with advanced AI features. Ensuring secure, compliant handling of sensitive medical data.
Accomplishments that we're proud of:
Designed a predictive donor model that works even on sparse datasets. Created a seamless end-to-end blood matching demo. Built a functional doctor-patient chat module within hackathon time. Presented a full concept that is scalable, integrable, and ready for pilot launch.
What we learned:
Real-world medical tech must prioritize data privacy and ease of access. AI can truly add value when used to augment existing efforts (like Blood Warriors) rather than replace them. The importance of designing inclusive tech that even non-tech-savvy users can benefit from.
What's next for ThalCare AI:
Integrate real datasets from Blood Warriors and hospitals for better prediction accuracy. Collaborate with e-RaktKosh for direct system linkage. Expand multilingual education modules for regional accessibility. Launch a pilot in partnership with Blood Warriors in selected Indian cities. Explore use of federated learning for AI to protect user privacy.
Built With
- aws3
- dart
- e-raktosh
- express.js
- firebase
- flutter
- gcp
- google-maps
- javascript
- jwt
- mongodb
- python
- react.js
- scikit-learn
- socket.io
- xgboost