esinet 0.1.1

Solves the M/EEG inverse problem using artificial neural networks with Python 3 and the MNE library.

esinet: Electric source imaging using artificial neural networks (ANNs)

esinet let's you solve the EEG inverse problem using ANNs. It currently supports one architecture: A fully connected neural network which is trained on single time instances of M/EEG data. This model was described in our paper.

---

Neural network design was created here

Dependencies:

• Python >= 3.8.3
• mne 0.22.0
  • Follow the installation guide
• Tensorflow>=2.4.1
  • Follow the installation guide
• Colorednoise
• joblib
• pyvista>=0.24
• pyvistaqt>=0.2.0
• tqdm

Installation from PyPi

Use pip to install esinet and all its dependencies from PyPi:

pip install esinet

Quick start

The following code demonstrates how to use this package:

from esinet import Simulation, Net

# Simulate M/EEG data
settings = dict(duration_of_trial=0.)
sim = Simulation(fwd, info, settings=settings)
sim.simulate(n_samples=10000)

# Train neural network on the simulated data
net = Net(fwd)
net.fit(sim)

# Perform predictions on your data
stc = net.predict(epochs)

First steps

Check out one of the tutorials to learn how to use the package:

• Tutorial 1: The fastest way to get started with the fully-connected ANN. This tutorial can be used as an entry point. If you want to dig deeper you should have a look at the next tutorials, too!
• Tutorial 2: Use esinet with low-level functions that allow for more control over your parameters with respect to simulations and training of the neural network.
• Tutorial 3: A demonstration of simulation parameters and how they affect the model performance.

Feedback

Please leave your feedback and bug reports at [e-mail].

Literature

Please cite us with this publication:

@ARTICLE{10.3389/fnins.2021.569918, AUTHOR={Hecker, Lukas and Rupprecht, Rebekka and Tebartz Van Elst, Ludger and Kornmeier, Jürgen},
TITLE={ConvDip: A Convolutional Neural Network for Better EEG Source Imaging},
JOURNAL={Frontiers in Neuroscience},
VOLUME={15},
PAGES={533},
YEAR={2021},
URL={https://www.frontiersin.org/article/10.3389/fnins.2021.569918},
DOI={10.3389/fnins.2021.569918},
ISSN={1662-453X} }

Troubleshooting

• Having problems with the installation? Check the package requirements

Notes on current version

• This version 0.1.0 is fully compatible with the mne-python package. This meant that I had to change the whole API to match the mne-python API. Please have a look at the new tutorials in order to get familiar with the new code structure or revert back to an earlier version. Check the changelog for a version list.