Sanjeevani AI

Inspiration

The inspiration behind Sanjeevani AI stems from the everyday struggles faced by Thalassemia patients and their families in India. Limited donor access, inconsistent blood availability, lack of awareness, and disconnected care systems make survival and quality of life extremely difficult. We wanted to create an AI-based, scalable solution that acts as a lifeline—one that connects patients with donors, hospitals, information, and support.

What it does

-Sanjeevani AI is a smart assistant and support platform for Thalassemia patients that: -Connects patients with potential blood donors using AI-based prediction -Provides a 24/7 AI chatbot for health-related queries and support -Maintains personalized treatment dashboards with reminders and history -Educates families and the public with NLP-driven learning modules -Ensures privacy while simplifying care journeys

How we built it

We built Sanjeevani AI using the following tools:

1. Microsoft Azure Cognitive Services – for chatbot, language translation, and image processing
2. Azure Bot Framework – for the CareBot assistant
3. Python & Scikit-learn – for the donor prediction model
4. Azure Functions & Blob Storage – for serverless backend and file handling
5. Firebase / Azure Cosmos DB – for data storage
6. HTML, CSS, JS (optional frontend) – for the dashboard interface

Challenges we ran into

-Data availability: Accessing real-world, anonymized health or donor datasets was a challenge. -Privacy concerns: Handling sensitive medical information securely and responsibly. -Multilingual support: Adapting the educational content to multiple regional languages. -API integration: Limited availability of public APIs like Blood Bridge or e-RaktKosh.

Accomplishments that we're proud of

--Designed a complete AI-powered ecosystem for Thalassemia care. --Built a working chatbot prototype with contextual Q&A support. --Created a predictive model for donor availability using basic data inputs. --Addressed a socially impactful, underserved domain in India.

What we learned

1. Deep insights into the healthcare challenges faced by rare blood disorder patients
2. The importance of inclusive design and multilingual accessibility
3. Strategies to manage privacy and security in healthcare apps

What's next for Sanjeevani AI

-Integrate real-time blood bank APIs like e-RaktKosh -Launch a mobile app version for wider accessibility -Enhance the AI model with real donor and patient datasets -Add a mental health and community support module -Collaborate with hospitals, NGOs, and state governments to deploy in real use cases

Built With

• aiml
• and-blob-storage-ai/ml:-scikit-learn
• azure
• azure-functions
• bot-service
• cosmos
• css
• database:
• db
• e-raktkosh
• firebase
• html
• javascript
• js-(optional)-apis:-blood-bridge-(if-available)
• microsoft-azure:-cognitive-services
• python
• python-(for-donor-prediction)-frontend:-html
• scikit-learn