bioRxiv Science⌕ Search

Biology subjects

Huberty, S.

Publications and source records attributed to Huberty, S..

2 recordsLinked to original sources

Deep learning and eye-tracking for accurate EOG rejection

The electroencephalogram (EEG) directly measures the electrical activity generated by the brain. Unfortunately, it is often contaminated by various artifacts, notably those caused by eye movements and blinks (EOG artifacts). Such artifacts are usually removed using an independent component analysis (ICA) or other blind source separation techniques. However, it is difficult to assess whether subtracting EOG components estimated through ICA removes some neurogenic activity. It is crucial to address this question to avoid biasing EEG analyses. Toward that objective, we developed a deep learning model for EOG artifact removal that exploits information about eye movements available through eye-tracking (ET). Using a multimodal EEG and ET open-access dataset, we trained within-subject a long short-term memory (LSTM) model to predict the component of EEG signals predictable from ET data. We further used this ET-informed evaluation of EOG artifacts to investigate the sensitivity and specificity of ICA. Our analysis indicates that although ICA is very sensitive to EOG, it has a comparatively low specificity. These results motivate further research on EEG artifact removal to develop approaches with higher EOG rejection specificity.

neuroscience↗

PyLossless: A non-destructive EEG processing pipeline

EEG recordings are typically long and contain large amounts of data, making manual cleaning a time-consuming and error-prone task. Automated pre-processing pipelines can facilitate the efficient and objective extraction of artifacts, enabling standardized and reproducible analyses. However, automated pre-processing pipelines typically remove data considered artifact, and return a subset of irreversibly transformed signals. This approach obfuscates pre-processing decisions, and often makes it impossible to recover the original data or modify the pre-processing steps. Further, it complicates collaboration between research teams working on a common dataset, because different analyses may require specific pre-processing routines. Given the large amount of resources that are devoted to collecting EEG, tools that can help efficiently and transparently pre-process data are greatly needed. PyLossless addresses this need by creating a non-destructive, automated pre-processing pipeline that maintains the continuous EEG structure. It offers a user-friendly API, it is well documented, tested through continuous integration, easily deployable, and integrates with the popular MNE-Python environment. The pipeline further provides a browser-based quality control review (QCR) dashboard that allows researchers to visualize and edit the automated artifact annotations on sensors, time-periods, and independent components. The end product of PyLossless is a lossless annotated data state that can be shared and used with analysis-specific artifact rejection policies, allowing for an optimal balance between flexibility and standardization.

neuroscience↗