Smart Med Tutor is an intelligent AI-powered medical assistant built using LangGraph. It is designed to support students across a range of biology and medical education levels — from high school to professional programs like MBBS, BDS, Pharmacy, and MDCAT preparation.
This assistant combines a Retrieval-Augmented Generation (RAG) approach using ChromaDB embeddings with natural LLM-based conversation to deliver accurate, simplified, and context-aware responses to complex medical queries.
| Layer | Technology |
|---|---|
| 🤖 Agent Framework | LangGraph |
| 🧠 Language Model | Any LLM via API (Tested with Mistral 7B on OpenRouter) |
| 📚 Embedding Store | ChromaDB for fast retrieval |
| 📄 Dataset | MedMCQA medical question-answer corpus |
| 🎨 UI Framework | Streamlit for interactive frontend |
| 📦 Vector Embedding | Sentence Transformers (all-MiniLM-L6-v2) |
Smart Med Tutor is tailored for:
-
📘 High School Students (Grades 9–12)
Strengthen your understanding of biology and related sciences. -
🩺 Pre-Medical Students (MDCAT, NEET, etc.)
Practice concept-based and clinical MCQs with quick explanations. -
🧠 MBBS / BDS / Nursing Students
Get support with medical theory, pharmacology, pathology, and case-based reasoning. -
💊 Pharmacy & Allied Health Students
Ask drug-related, dosage-based, or scenario-based questions and get clear responses.
Smart Med Tutor can handle a wide range of questions, including but not limited to:
- Conceptual questions (e.g., "How do enzymes work in digestion?")
- Clinical questions (e.g., "A patient with chest pain and sweating, what could it indicate?")
- Drug-related queries (e.g., "What are the uses and side effects of Augmentin?")
- Medical MCQs and reasoning
- Biology theory from class 9th to university-level
- Medical entrance exam practice
- Pharmacology, pathology, anatomy, and physiology topics
Smart Med Tutor uses intelligent question understanding combined with a medical knowledge base derived from real-world datasets like MedMCQA. Responses are generated using advanced AI models for clarity and accuracy.
We believe in supporting students, educators, and developers working in healthcare and edtech. While the frontend app is available for demonstration, the full working project including the backend logic, embeddings, and AI agent pipeline is available through collaboration.
If you are:
- An educator or tutor building student tools
- A developer working on health education apps
- A researcher working in AI for medical education
➡️ We'd love to connect and explore how Smart Med Tutor can help your goals.
For feedback, collaboration, or partnership opportunities:
- 💼 Email: syedsaadi427@gmail.com
- 🌐 GitHub: @syedsaadali11
Smart Med Tutor is your companion in mastering medicine, biology, and health sciences — designed for the next generation of healthcare professionals.