🧠 Project: MediAssistAI 🌐 Live Website
In rural India, access to quality healthcare is often hindered by limited infrastructure, lack of medical professionals, and low literacy. MediAssistAI is built with the mission to bridge this gap using the power of Artificial Intelligence.
By combining voice-enabled symptom input, image-based diagnosis, and a comprehensive symptom checker, SehatAI empowers users—especially from rural and underserved areas—to understand their health conditions better. In addition, it provides educational resources and information on government health schemes to spread awareness and encourage healthcare engagement.
MediAssistAI is your intelligent village health assistant—reliable, accessible, and inclusive.
MediAssistAI is a rural-friendly AI-based diagnostic tool that allows users to:
- ✅ Check symptoms by selecting age, gender, severity, and specific symptoms
- ✅ Upload medical images (e.g., X-rays) for ML-powered disease analysis
- ✅ Use voice input to describe symptoms in natural language
- ✅ Explore government healthcare schemes tailored to their needs
- ✅ Gain health awareness through educational resources
🎯 Accuracy: Our ML models achieve up to 90% accuracy in symptom-based and image-based disease prediction, delivering reliable insights to aid early detection.
This tool provides a smart, multimodal, and inclusive health assistant experience.
- React.js
- Tailwind CSS
- JavaScript
- Axios
- Flask (Python)
- Machine Learning Models
- Speech Recognition (for voice input)
- Image Classification (for medical images)
- RESTful APIs
- CORS
- JSON-based communication
- Git & GitHub
- Postman
- Render / Railway (for backend)
- Netlify / Vercel (for frontend)
The UI is built using React with Tailwind CSS for styling. It’s designed to be lightweight and mobile-friendly for rural areas with limited internet connectivity. Users can input symptoms via text or voice and navigate the app easily.
The Flask backend handles requests for:
- Symptom-to-disease prediction using trained ML models.
- Image upload and classification.
- Voice-to-text conversion for symptom input.
- Fetching government scheme data.
The ML models are trained on datasets for diseases like pneumonia, tuberculosis, and common viral infections. The system provides an accuracy of ~90%, ensuring reliability in its predictions.
- ✅ Developed a multi-input health diagnostic tool (text, image, voice)
- ✅ Achieved ~90% accuracy in diagnosis
- ✅ Optimized for rural areas: minimal data usage, simple UI
- ✅ Integrated government health scheme data dynamically
- ✅ Created awareness tools to empower users with health education
Follow these steps to run MediAssistAI locally:
git clone https://github.com/shwetaverma14/aixplain.git
cd aixplaincode .cd backend
python -m venv venv
source venv/bin/activate
pip install -r requirements.txt
python app.py
cd client
npm install
npm run dev
-
🌐 Local language support:
Add multilingual support including Hindi and regional dialects to improve accessibility for rural users. -
📱 Android app version:
Develop an offline-first mobile application for areas with limited internet connectivity. -
📊 Health trend analytics:
Enable NGOs and healthcare professionals to view anonymous health trends, supporting better policy planning. -
🤖 Conversational AI Integration:
Integrate ChatGPT or similar LLMs for a chat-based medical assistant experience, making interactions more intuitive.
B.Tech CSE | Full Stack Developer
🔗 GitHub Profile
B.Tech CSE | Full Stack Developer
🔗 GitHub Profile
🙌 Thanks for checking out MediAssistAI!