AI-Medical-Research-Question-Answering-System

AI - Medical Research QA System : Project Objective This project aims to create an intelligent question-answering system specifically designed for medical research. It leverages advanced RAG (Retrieval Augmented Generation) technology to provide accurate, research-backed responses to medical queries by accessing and analyzing a comprehensive database of medical research papers and literature.

Task

Built RAG (Retrieval-Augmented Generation) system that utilizes Ollama LLM to answer medical research queries with source citations.

Features

• Vector-based search using ChromaDB
• Natural language query interface with Streamlit
• Source citation for answers
• Semantic matching using Sentence Transformers

Prerequisites

• Python 3.8+
• Ollama (install from ollama.ai)
• 8GB RAM minimum
• 20GB disk space

Tech stack

Installation and Usage

# Step1:  Clone repo
git clone https://github.com/Modupeolawuraola/AI-Medical-Research-Question-Answering-System

# Step 2: Create a virtual environment and setup the virtual environment with the command below
python -m venv -rag_env
source rag_env/bin/activate

# Step3: Install dependencies
pip install -r requirements.txt

#Step 4: download ollama locally on your computer  using the link below
install ollama : https://ollama.com/download/windows

# Step5: Pull Llama2 model
ollama pull llama2

# Step6: re-run the setuptool -setup.py and pyproject.toml tool by running the command below
pip3 install  -e .

#step7: Run the test code to confirm the app is running effectively with the command below
python test_rag.py

# Step 8: usage
streamlit run streamlit_app.py

# Step 9: Using docker to host the app locally on your desktop 
install docker desktop on your mac or windows system 

verify the installation by running this command 
docker --version

once its install build your docker  by running the command below 
docker build -t medical-rag-app -f src/rag/ui/docker/Dockerfile .

run docker with the command below 
docker run -p 8501 mdeical-rag-app (for port 8501)
docker run -p 8502:8501 medical-rag-app( for non port 8501)

NOTE THIS APP can only run locally on your computer except you want to deploy and host on aws ec2 or gcp or digital ocean which cost money 

App Screenshot

Who Can Use This System?

Healthcare Professionals

• Medical Doctors seeking quick access to research findings
• Medical Researchers looking to explore existing literature
• Healthcare Students studying medical research papers
• Clinical Trial Coordinators needing to reference similar studies

Academic Community

• Medical School Students
• Research Scholars in healthcare fields
• University Professors teaching medical courses
• PhD Candidates conducting literature reviews

Research Organizations

• Medical Research Institutes
• Pharmaceutical Companies
• Healthcare Technology Companies
• Clinical Research Organizations

Healthcare Writers and Communicators

• Medical Journal Editors
• Healthcare Content Writers
• Medical Documentation Specialists
• Health Education Content Creators

Benefits for Users

• Quick access to relevant medical research information
• Evidence-based responses backed by research papers
• Time-saving alternative to manual research
• Ability to explore complex medical topics efficiently
• Access to comprehensive medical literature analysis

Folder structure

medical-research-qa-rag/
├── README.md
├── assets/
├── docs/
├── rag_env/ ✗[.gitignore]
├── src/
│   ├── medical_rag.egg-info/ ✗[.gitignore]
│   └── rag/
│       ├── ui/
│       │   └── streamlit_rag_app/
│       ├── __init__.py
│       ├── database.py
│       ├── embedding.py
│       ├── preprocessing.py
│       ├── retrieval.py
│       └── utils.py
├── tests/
│   ├── data/ ✗
│   ├── medical_research_db/ ✗[.gitignore]
│   └── test_rag.py
├── .gitignore
├── main.py
├── pyproject.toml
├── requirements.txt
└── setup.py