Critical_PCI

This repository contains all code used for the publication: "Critical dynamics in spontaneous EEG predict anesthetic-induced loss of consciousness and perturbational complexity"

The code was written by Charlotte Maschke and Jordan O'Byrne

requirements

pip install -r requirements.txt also dependent on EdgeofPy toolbox (https://github.com/jnobyrne/edgeofpy)

Available Features

the input parameters of all functions can be seen using the comand python functionNAME.py --help

Most functions output a .txt with all features over subjects and conditions. Some scripts additionally output space-resolved features per subject.

• Avalanche Criticality features

• features_AVC.py

  • alpha, tau, third
  • deviation from criticality coefficient
  • branching ratio
  • repertoire
  • intrinsic timescale (not validated and not used in the paper)
  • susceptibility (not validated and not used in the paper)

• features_avc_std_dist.py

  • distribution of Avalances over standard deviation
  • amount of Avalanches

• Edge of Chaos features

• features_EOC.py

  • 01 Chaos test (different options, see --help)

• Edge of Synchrony features

• features_EOS.py

  • Pair Correlation Function (different options, see --help)
  • Order parameter

• Criticality-related features

• features_Comp.py

  • Lempel-Ziv complexity univariate, shuffle and length normalized
  • Fractal dimension

• features_DFA.py

  • Hurst exponent

• features_slope.py

  • Spectral Slope

• Other features

• features_power.py

  • spectral power in frequency bands

METHODS

Herein used utility functions METHODS_chaos and METHODS_EOS are now part of functions of EdgeofPy.