Ovarian Cancer Chatbot

Overview

This chatbot leverages Mistral 7B and LangChain to provide ovarian cancer-related information and calculate ROMA scores based on biomarker inputs. It can also answer general queries and extract relevant data from PDFs.

Features

• ROMA Score Calculation: Risk assessment using CA125 and HE4 biomarkers.
• PDF Querying: Retrieve information from stored PDFs on hospitals, diet plans, and ovarian cancer.
• General Health Chat: Responds to general queries using HuggingFace models.

##ROMA Score Calculation

• Postmenopause

  ROMA = (exp(-12 + 2.38 * ln(HE4) + 0.0626 * ln(CA125)) / (1 + exp(-12 + 2.38 * ln(HE4) + 0.0626 * ln(CA125)))) * 100

• Pre-menopause

  ROMA = (exp(-8.09 + 1.04 * ln(HE4) + 0.732 * ln(CA125)) / (1 + exp(-8.09 + 1.04 * ln(HE4) + 0.732 * ln(CA125)))) * 100

Tech Stack

• HuggingFace Hub: Hosts the Mistral 7B model.
• LangChain: Manages PDFs and FAISS-based vector search.
• Python & Flask: Backend for chat handling.

Installation

1. Clone the repository:

   git clone https://github.com/thanish11/Langchain_HuggingFaceHub.git
   cd Langchain_HuggingFaceHub