Codebase of the WOCCA algorithm

This codebase contains the implementation of the WOCCA algorithm proposed in the manuscript "Spatiotemporal Decomposition of Whole-Brain Alpha Traveling Waves", supplementary materials of this manuscript, and notebooks for reproduction of EEG dataset analysis, written in Python language with Jupyter Notebook.

Quick Start

For a quick start on how to use WOCCA, refer to grid_samples.py for examples on synthetic grid data. For reproduction of analysis and creation of figures in the manuscript, follow the instructions in the data directory to set up the EEG datasets, before referring to the Jupyter Notebooks.

Requirements

mne
torch
scipy
statsmodels
scikit-learn
pandas
matplotlib

Contents

The core WOCCA algorithm

• wocca.py: Main implementation of the WOCCA algorithm

Supporting Code

• core.py: Utilities and definition of data structures
• gradient.py: Tools for processing phase gradient field
• grid_samples.py: Tools for synthetic data
• icons.py: Draw various icons for figures in the manuscript
• ring.py: Experimental dynamical analysis tools, only used for generating geodesic interpolation of synthetic data
• templates.py: Microstate analysis
• visualize.py: Various visualization tools

Datasets

• ds1_leipzig.py
• ds2_trt.py
• ds3_sedation.py
• preprocess_ds2_trt.py: Manual preprocessing pipeline for Dataset 2

Analysis and Figures

• nb1_explore_ds1_leipzig.ipynb: Exploration of Dataset 1 (Leipzig)
• nb2_explore_ds2_trt.ipynb: Exploration of Dataset 2 (TRT)
• nb3_explore_wocca_consistency.ipynb: Robustness analysis between Dataset 1 and 2
• nb4_explore_microstate.ipynb: Microstate analysis of Dataset 1
• nb5_explore_ds3_sedation.ipynb: Exploration of Dataset 3 (Sedation)
• nb6_intro_and_toy_data.ipynb: Introductory diagrams and synthetic data analysis

Supplementary information

• supplementary_information.pdf: Supplementary details of mathematics, statistics and data analysis