Medbot

Enhancing diagnosis ofrespiratory diseases through machine learning and generative AI .

Key Features • How To Use • Technical Skills Gained • License

Key Features

• Authentication and authorization

  • Create your own account and log in

• Chat-based consultation

  • The app answers questions about Tuberculosis and Pneumonia following RAG principles

• Image Classification

  • Contains a DenseNet-121 computer vision model to classify chest X-ray images into three categories: Normal, Tuberculosis, and Pneumonia. *Adherence to responsible computing aspects
  • Chat agent adheres to a meta prompt that limits its capabilities to asnwering questions about Pneumonia and Tuberculosis
  • Has an application of the LIME Python library to enhance computer vision's model interpretability

How To Use

To clone and run this application, you'll need Git and Node.js (which comes with npm) Python and Flask installed on your computer. From your command line:

# Clone this repository
$ git clone https://github.com/stephmukami/medbot.git

# Go into the repository
$ cd client2

# Install dependencies
$ npm install

# Run the app
$ npm start

# Run the ml backend
$ newenv\Scripts\activate
$ cd server
$ uvicorn main:app --host 0.0.0.0 --port 7000
$ newenv\Scripts\deactivate 

Note If you're using Linux Bash for Windows, see this guide or use node from the command prompt.

License

MIT

Technical Skills Gained

• Integration of Vector Databases ie Qdrant and Pinecone
• Data scraping using Beautiful Soup
• RAG architecure
• Intergration of LLMs ie Open AI
• API design using FastAPI
• Responsive web design
• Use of ORMs and Postgres as a service ie Prisma & Railway
• Testing using PyTest and RAGAs (Retrieval-Augmented Generation Assessment)

To clone and run this application, you'll need Git and Node.js (which comes with npm) Python and Flask installed on your computer. From your command line:

---

Dream Team :

Stephanie Mukami

GitHub @stephmukami  · 

Rose Kimu

GitHub @Rose-Kimu  ·