v1.0.0

Release v1.0.0

Overview

This repository contains the code for the Data Science Challenge problem on machine learning for cardiac electrophysiology. It includes comprehensive tutorials on machine learning and a set of starter notebooks to help you dive into the challenge. These notebooks are designed as a launching pad—you are encouraged to modify and extend them using any machine learning framework (e.g., PyTorch, TensorFlow, JAX, etc.) and model of your choice (e.g., neural networks, random forests, etc.).

Key Features

• Tutorials & Notebooks: Learn and experiment with machine learning techniques tailored for cardiac electrophysiology.
• Framework Agnostic: Implement solutions with your preferred machine learning framework.
• Helper Functions: Easily load data from the Dataset of Simulated Intracardiac Transmembrane Voltage Recordings and ECG Signals.
• Extensible Codebase: Built as a starting point for further exploration and development.

References & Resources

• Used for 2023 and 2024 LLNL Data Science Challenge:
  • LLNL Data Science Class 2023
  • LLNL Data Science Class 2024
• Inspiration & Background:
  • Medium Blog post by Mikel Landajuela provides useful insights to get started.
  • Code implementation is based on the cardiac_ml repository

Getting Started

Clone the repository, explore the tutorials and notebooks, and customize them according to your project requirements. Enjoy experimenting with different machine learning models and frameworks to tackle the challenge of cardiac electrophysiology!