In Vivo Tumour Mapping Using Electrocorticography Alterations During Awake Brain Surgery: A Pilot Study

Decoding Intracranial EEG With Machine Learning: A Systematic Review

Advances in intracranial electroencephalography (iEEG) and neurophysiology have enabled the study of previously inaccessible brain regions with high fidelity temporal and spatial resolution. Studies of iEEG have revealed a rich neural code subserving healthy brain function and which fails in disease states. Machine learning (ML), a form of artificial intelligence, is a modern tool that may be able to better decode complex neural signals and enhance interpretation of these data. To date, a number of publications have applied ML to iEEG, but clinician awareness of these techniques and their relevance to neurosurgery, has been limited. The present work presents a review of existing applications of ML techniques in iEEG data, discusses the relative merits and limitations of the various approaches, and examines potential avenues for clinical translation in neurosurgery. One-hundred-seven articles examining artificial intelligence applications to iEEG were identified from 3 databases. Clinical applications of ML from these articles were categorized into 4 domains: i) seizure analysis, ii) motor tasks, iii) cognitive assessment, and iv) sleep staging. The review revealed that supervised algorithms were most commonly used across studies and often leveraged publicly available timeseries datasets. We conclude with recommendations for future work and potential clinical applications.

Time rescaling reproduces EEG behavior during transition from propofol anesthesia-induced unconsciousness to consciousness

General anesthesia (GA) is a reversible manipulation of consciousness whose mechanism is mysterious at the level of neural networks leaving space for several competing hypotheses. We recorded electrocorticography (ECoG) signals in patients who underwent intracranial monitoring during awake surgery for the treatment of cerebral tumors in functional areas of the brain. Therefore, we recorded the transition from unconsciousness to consciousness directly on the brain surface. Using frequency resolved interferometry; we studied the intermediate ECoG frequencies (4-40 Hz). In the theoretical study, we used a computational Jansen and Rit neuron model to simulate recovery of consciousness (ROC). During ROC, we found that f increased by a factor equal to 1.62 ± 0.09, and δf varied by the same factor (1.61 ± 0.09) suggesting the existence of a scaling factor. We accelerated the time course of an unconscious EEG trace by an approximate factor 1.6 and we showed that the resulting EEG trace match the conscious state. Using the theoretical model, we successfully reproduced this behavior. We show that the recovery of consciousness corresponds to a transition in the frequency (f, δf) space, which is exactly reproduced by a simple time rescaling. These findings may perhaps be applied to other altered consciousness states.