Classification of Movement-Related Oscillations in sEEG Recordings with Machine Learning

Abstract

Previous electrophysiological research has characterized canonical oscillatory patterns associated with movement mostly from recordings of primary sensorimotor cortex. Less work has attempted to decode movement based on electrophysiological recordings from a broader array of brain areas such as those sampled by stereoelectroencephalography (sEEG). Here we decoded movement using a linear support vector machine (SVM). We were able to accurately classify sEEG spectrograms during a keypress movement in a task versus those during the inter-trial interval. Furthermore, the important time-frequency patterns for this classification recapitulated findings from previous studies that used non-invasive electroencephalography (EEG) and electrocorticography (ECoG) and identified brain regions that were not associated with movement in previous studies. Specifically, we found these previously described patterns: beta (13 - 30 Hz) desynchronization, beta synchronization (rebound), pre-movement alpha (8 - 15 Hz) modulation, a post-movement broadband gamma (60 - 90 Hz) increase and an event-related potential. These oscillatory patterns were newly observed in a wide range of brain areas accessible with sEEG that are not accessible with other electrophysiology recording methods. For example, the presence of beta desynchronization in the frontal lobe was more widespread than previously described, extending outside primary and secondary motor cortices. We provide evidence for a system of putative motor networks that exhibit unique oscillatory patterns by describing the anatomical extent of the movement-related oscillations that were observed most frequently across all sEEG contacts. Significance StatementSeveral major motor networks have been previously delineated in humans, however, much less is known about the population-level oscillations that coordinate this neural circuitry, especially in cortex. Therapies that modulate brain circuits to treat movement disorders, such as deep brain stimulation (DBS), or use brain signals to control movement, such as brain-computer interfaces (BCIs), rely on our basic scientific understanding of this movement neural circuitry. In order to bridge this gap, we used stereoelectroencephalography (sEEG) collected in human patients being monitored for epilepsy to assess oscillatory patterns during movement.