Skip to content

AIDx Model Development: Developing AI-driven models for healthcare applications.

License

Notifications You must be signed in to change notification settings

AIDx-Med/AIDx-Model-Development

Repository files navigation

AIDx Model Development

Overview 📖

AIDx Model Development is an innovative repository focused on developing AI-driven models for healthcare applications. This project specializes in processing medical data, particularly from the MIMIC-IV database, and offers tools for model training and dataset management.

Features 🚀

  • Data Processing: Scripts for tokenizing and transforming healthcare data such as MIMIC-IV 🏥.
  • Model Training & Fine-Tuning: Utilizes fine-tuning to create AI models on medical datasets 🧬.
  • Dataset Management: Tools to create and handle datasets, including Parquet datasets, for healthcare AI 📊.

Components 🛠

  • aidx.py: The gateway to various functionalities like dataset creation and model tuning 🌐.
  • Docker Support: Facilitates hosting the Jupyter Notebook environment 🐳.
  • Notebooks: Step-by-step guides in Jupyter notebooks for data processing and analysis 📓.
  • Scripts: Python scripts for comprehensive processes including database interactions and data workflows 📜.
  • Dependencies: requirements.txt for easy installation of necessary packages 📌.

Other Requirements

  • Google Drive and rclone for dataset storage
    • Make sure the rclone.conf file is in the scripts/utils directory
  • MIMIC-IV Database with PostgreSQL

Getting Started 🚀

  1. Clone the repository.
  2. Set up the Docker environment, which hosts the Jupyter Notebook.
  3. Install dependencies from requirements.txt.
  4. Run python aidx.py to access the command menu and start exploring the functionalities.