ThalAI

We would like to have our interface in this way.

ThalAI – An AI-Powered Ecosystem for Thalassemia Care

Inspiration Thalassemia affects thousands of lives across India, especially in under-resourced regions where regular access to blood and care is limited. We were inspired by the stories of "Blood Warriors" who relentlessly support these patients and wanted to build a scalable, intelligent platform that would make their mission easier, faster, and more secure.

The lack of awareness, fragmented donor systems, and overburdened caregivers motivated us to create ThalAI — a one-stop solution that leverages technology for impact.

What it does ThalAI is a secure, AI-driven web platform that:

Connects Thalassemia patients with compatible blood donors in real-time
Predicts donor availability using past donation patterns and health data
Provides a multilingual chatbot for education and support
Ensures secure data handling using encrypted storage (and optional blockchain)
Integrates with e-RaktKosh and Blood Bridge APIs for seamless coordination

How we built it We divided our development into functional modules:

Frontend: React.js + TailwindCSS for responsive UI
Backend: Node.js + Express.js for REST APIs
AI Model: Python with Scikit-learn (logistic regression and time-series model) to predict donor availability
Chatbot: Rasa for NLP, trained on FAQs and educational content
Database: PostgreSQL with AES-256 encryption
Integration: Mock endpoints for e-RaktKosh and Blood Bridge (ready for production API keys)
Hosting: Vercel (frontend) and Render.com for backend

We used tools like Figma for wireframes, and GitHub for version control and team collaboration.

Challenges we ran into

Designing an AI model that accounts for irregular donor behavior while respecting donation eligibility rules (e.g., 90-day gaps)
Creating a multilingual, context-aware chatbot that could educate users while also responding to emotional queries
Ensuring data privacy and consent mechanisms without slowing down the user experience
Simulating integrations with national blood systems without access to real-time APIs during the hackathon

Accomplishments that we're proud of

Developed an end-to-end working prototype within the hackathon timeframe
Successfully built and tested an AI model that predicts donor availability with >85% accuracy
Integrated a working multilingual chatbot using Rasa, covering over 50 FAQs and care instructions
Created a secure user management system with role-based access control
Designed an architecture ready to scale with government/NGO integrations

What we learned

How to break a large social-healthcare problem into scalable technical modules
The importance of UX when designing for patients and caregivers who may not be tech-savvy
Real-world limitations of blood donation systems and how AI can optimize them without risking safety
Designing for empathy — especially in cases where users are emotionally and physically vulnerable

What's next for ThalAI