This Jupyter Notebook provides a step-by-step walkthrough of the code. It includes eyeball coregistration, voxel extraction, model training and test as well as basic performance measures. Alternatively, here is a Colab Notebook that runs in the browser. DeepMReye also includes a user-friendly streamlit app (installation option 4): Drag and drop your fMRI data into the interface and download your gaze coordinates shortly after.

This Data Repository includes exemplary data for model training and test, source data of all paper figures as well as pre-trained model weights.

Moreover, here are additional User Recommendations as well as a Frequently-Asked-Questions (FAQ) page. If you have other questions, please reach out to us.

Install DeepMReye with a CPU/GPU version of TensorFlow using the following command.

pip install deepmreye

To encapsulate DeepMReye in a virtual environment install with the following commands:

conda create --name deepmreye python=3.9
conda activate deepmreye
pip install deepmreye

For GPU support, follow tensorflow install instruction, e.g.:

conda install -c conda-forge cudatoolkit=11.2 cudnn=8.1.0

If installation of ANTsPy fails try to manually install it via:

git clone https://github.com/ANTsX/ANTsPy
cd ANTsPy
pip install CMake
python3 setup.py install

We provide a Colab Notebook showcasing model training and evaluation on a GPU provided by Google Colab. To use your own data, preprocess your data locally and upload only the extracted eyeball voxels. This saves space and avoids data privacy issues. See the Jupyter Notebook for the preprocessing and eyeball-extraction code.

Pull the image from docker hub.

docker pull deepmreye/deepmreye

Use deepMReye in a docker container via jupyterlab:

mkdir -p $PWD/notebooks
docker run -it --rm \
    --publish 8888:8888 \
    --volume $PWD/notebooks:/home/neuro/notebooks \
    deepmreye/deepmreye:latest \
        jupyter-lab --no-browser --ip 0.0.0.0

If you would like decode gaze coordinates in your data using a pretrained model, the easiest way is using our streamlit app.

Running the following commands will open a browser window that allows you to upload your data (.nii or .nii.gz) and then download the corresponding gaze coordinates shortly after. Please read the User Recommendations before using the pretrained models.

git clone https://github.com/DeepMReye/DeepMReye.git
cd DeepMReye
pip install .
pip install streamlit
streamlit run streamlit/streamlit.py

The fMRI data should be organized in 4D NIFTI files (.nii), containing the realigned 3D images acquired over time. The pipeline then extracts the eyeball voxels automatically and saves them as Python Pickle files, which serve as model input. For model training, you additionally need training labels, a numpy array containing 10 gaze coordinates per functional volume. These gaze coordinates can either be camera-based eye-tracking labels or the coordinates of a fixation target, and many file formats can be easily read (e.g. .npy, .npz, .mat, .csv etc.).

Please see our FAQ page for more details on data formats and preprocessing.

The GPU version of DeepMReye requires a NVIDIA GPU.

The following python dependencies are being automatically installed when installing DeepMReye (specified in setup.cfg):

tensorflow-gpu (2.2.0)
numpy (1.19.1)
pandas (1.0.5)
matplotlib (3.2.2)
scipy (1.5.0)
ipython (7.13.0)
plotly (4.14.3)

Version in parentheses indicate the ones used for testing the framework. Its extensively tested on Linux 16.04 but should run on all OS (Windows, Mac, Linux) supporting a Python version >3.8 and pip. It is recommended to install the framework and dependencies in a virtual environment (e.g. conda).

If you would like to run a pretrained version of DeepMReye on datasets that follow the Brain Imaging Data Structure (BIDS) format, check out bidsMReye: a wrapper for DeepMReye for BIDS datasets (incl. fMRIprep outputs) developed by Remi Gau. Please read the paper and User Recommendations before you use it.

If you have questions, comments or inquiries, please check out the online User documention and reach out to us: markus.frey[at]ntnu.no & matthias.nau[at]ntnu.no