MediBot

MediBot is a suite of advanced medical chatbots designed to assist with various healthcare tasks. The project is organized into three main applications: Basic, Report Assistant, and Advanced, each providing specialized functionalities for different use cases.

Table of Contents

• Introduction
• Demo Video
• Installation
• Usage
• Features
• Dependencies
• Configuration
• Contributors

Introduction

MediBot is designed to offer medical assistance through interactive chat interfaces. It utilizes OpenAI's GPT-4 model to provide responses and facilitate conversations tailored to medical contexts. The project includes:

• Basic: A simple chat interface for basic medical queries.
• Report Assistant: Assists with report analysis and answers questions based on provided documents.
• Advanced: A comprehensive medical diagnostic system with multiple agents and tasks.

Demo-Video

Installation

To set up the MediBot, follow these steps:

1. Clone the repository:

   git clone https://github.com/yourusername/MediBot.git
   cd MediBot

2. Create and activate a virtual environment:

   python3 -m venv env
   source env/bin/activate

3. Install the dependencies:

   pip install -r requirements.txt

4. Set up environment variables:

   • Create a .env file in the project root directory.
   • Add your OpenAI API key:

     OPENAI_API_KEY=your_openai_api_key
     

Usage

Basic

To run the Basic MediBot:

panel serve src\basic.py

Report Assistant

To run the Report Assistant MediBot:

panel serve src\report_assistant.py

Advanced

To run the Advanced MediBot:

panel serve src\advanced.py

All 3 together

To run all 3 together and get an option to select from these three:

panel serve src\advanced.py src\basic.py src\report_assistant.py

Then go to your browser and type:

http://localhost:5006/

Features

Basic:

• Simple chat interface
• Real-time responses using OpenAI's GPT-4 model

Report Assistant:

• Upload PDF reports for analysis
• Retrieve answers to queries based on the content of the reports

Advanced:

• Multi-agent system with specific roles (Clinical Assistant, Medical Diagnostician, General Doctor)
• Sequential task processing for comprehensive diagnostics
• Customizable human interface for interaction

Dependencies

openai
panel
dotenv
langchain
chroma
crew
crewai

Configuration

YAML Configuration Files

Agents Configuration (config/agents.yaml): Defines the agents involved in the diagnostic process.

Tasks Configuration (config/tasks.yaml): Specifies the tasks and the sequence of their execution.

Contributors

• Mudit Agrawal
• Mridul Gupta