Biology subjects

Diachenko, M.

Publications and source records attributed to Diachenko, M..

2 recordsLinked to original sources

bioRxiv2024

Qualitative EEG abnormalities signal a shift towards inhibition-dominated brain networks. Results from the EU-AIMS LEAP studies

Qualitative EEG abnormalities are common in Autism Spectrum Disorder (ASD) and hypothesized to reflect disrupted excitation/inhibition balance. To test this, we recently introduced a functional measure of network-level E/I ratio (fE/I). Here, we applied fE/I and other EEG measures to alpha oscillations from source-reconstructed data in the EU-AIMS dataset (267 ASD, 209 controls). We analyzed these measures alongside qualitative EEG abnormalities ranging from slowing of activity to epileptiform patterns, aiming to replicate the findings from the SPACE-BAMBI study. EEG abnormalities were rare in adults and could not be statistically assessed. ASD children-adolescents with EEG abnormalities exhibited lower relative alpha power and fE/I compared to those without. However, EEG-abnormality scoring did not stratify the behavioral heterogeneity of ASD using clinical measures. Surprisingly, several controls also exhibited qualitative EEG abnormalities with a strikingly similar anatomical distribution of reduced fE/I, suggesting a shift towards inhibition-dominated network dynamics in sensory processing regions. The robustness of this association between EEG abnormalities and reduced fE/I was further supported by re-analysis of the SPACE-BAMBI study in source space. Stratification by the presence of EEG abnormalities and their effects on network activity may help understand neurodevelopmental physiological heterogeneity and the difficulties in implementing E/I targeting treatments in unselected cohorts.

neuroscience↗

bioRxiv2022

Robin's Viewer: Using Deep-Learning Predictions to Assist EEG Annotation

Machine learning techniques such as deep learning have been increasingly used to assist EEG annotation, by automating artifact recognition, sleep staging, and seizure detection. In lack of automation, the annotation process is prone to bias, even for trained annotators. On the other hand, completely automated processes do not offer the users the opportunity to inspect the models output and re-evaluate potential false predictions. As a first step towards addressing these challenges, we developed Robins Viewer (RV), a Python-based EEG viewer for annotating time-series EEG data. The key feature distinguishing RV from existing EEG viewers is the visualization of output predictions of deep-learning models trained to recognize patterns in EEG data. RV was developed on top of the plotting library Plotly, the app-building framework Dash, and the popular M/EEG analysis toolbox MNE. It is an open-source, platform-independent, interactive web application, which supports common EEG-file formats to facilitate easy integration with other EEG toolboxes. RV includes common features of other EEG viewers, e.g., a view-slider, tools for marking bad channels and transient artifacts, and customizable preprocessing. Altogether, RV is an EEG viewer that combines the predictive power of deep-learning models and the knowledge of scientists and clinicians to optimize EEG annotation. With the training of new deep-learning models, RV could be developed to detect clinical patterns other than artifacts, for example sleep stages and EEG abnormalities.

neuroscience↗