Medical Chatbot System

Overview

Medical Chatbot System | Python, TensorFlow, scikit-learn, NLP | 2024

This repository implements an AI-driven medical chatbot system with three integrated modules for improving healthcare accessibility and preliminary support:

• ICD Recommender (LSTM): Predicts ICD codes from user-described symptoms using deep learning (TensorFlow).
• Healthcare Chatbot (Decision Tree): Provides preliminary diagnoses based on user symptom inputs.
• First Aid Bot (Pre-trained Models): Guides users with immediate first aid steps during emergencies.

Repository Structure

• main.py: Entry point for the Streamlit multi-page app, allowing users to navigate between the three chatbot modules seamlessly.
• icd.py, lstm.py: Implements the LSTM-based ICD code recommender.
• healthcare.py: Contains the healthcare chatbot using a decision tree for symptom-based diagnostics.
• first_aid.py: Contains the first aid guidance bot.
• EngChatbotModel.h5: Pre-trained model for language handling in chatbot interactions.
• health_image.jpeg: Visual asset for chatbot interface.
• templates/, static/, img/: Front-end assets for UI rendering in Streamlit.
• README.md: Documentation for this project.

Notes on Data

Please note that the datasets used for model training and evaluation have not been added to this repository due to size and privacy considerations. To replicate the workflow, you will need structured symptom and ICD datasets aligned with the system’s CSV-based ingestion pipeline.

Features

• Provides ICD code prediction using LSTM-based neural networks.
• Offers preliminary medical advice based on user symptoms.
• Delivers first aid guidance for emergency situations.
• Modular design, easily extendable with additional medical functionalities.
• Uses TensorFlow, scikit-learn, and NLP methods for robust medical text interpretation.
• Streamlit-based GUI for user-friendly interaction.

Getting Started

1. Clone the repository:

   git clone [email protected]:ArtsyInd/Medical-Chatbot.git
   cd Medical-Chatbot

2. Install required packages:

   pip install -r requirements.txt

3. Run the unified Streamlit app:

   streamlit run main.py

   Navigate through the ICD Recommender, Healthcare Chatbot, and First Aid modules via the sidebar.

License

This project is intended for academic and research learning purposes in healthcare AI development.

---

For any queries regarding usage or contributions, please feel free to raise an issue or discussion in this repository.