RT Journal Article
SR Electronic
T1 MR-based camera-less eye tracking using deep neural networks
JF bioRxiv
FD Cold Spring Harbor Laboratory
SP 2020.11.30.401323
DO 10.1101/2020.11.30.401323
A1 Frey, Markus
A1 Nau, Matthias
A1 Doeller, Christian F.
YR 2020
UL http://biorxiv.org/content/early/2020/12/01/2020.11.30.401323.abstract
AB Viewing behavior provides a window into many central aspects of human cognition and health, and it is an important variable of interest or confound in many fMRI studies. To make eye tracking freely and widely available for MRI research, we developed DeepMReye: a convolutional neural network that decodes gaze position from the MR-signal of the eyeballs. It performs camera-less eye tracking at sub-imaging temporal resolution in held-out participants with little training data and across a broad range of scanning protocols. Critically, it works even in existing datasets and when the eyes are closed. Decoded eye movements explain network-wide brain activity also in regions not associated with oculomotor function. This work emphasizes the importance of eye tracking for the interpretation of fMRI results and provides an open-source software solution that is widely applicable in research and clinical settings.Competing Interest StatementThe authors have declared no competing interest.