Identifying epileptogenic abnormality by decomposing intracranial EEG and MEG power spectra

Identifying abnormal electroencephalographic (EEG) activity is crucial in diagnosis and treatment of epilepsy. Improved EEG biomarkers of the epileptogenic zone are important, as around half of individuals have recurrent seizures after surgical treatment. In recent years, substantial research has focused on EEG biomarkers between epileptic seizures using normative maps of power in different frequencies. This approach showed promising results when comes to identification of abnormalities, however, it is unclear what spectral features drive them.

In this talk, I will describe how the decomposition of brain activity into periodic (oscillatory) and aperiodic (trend across all frequencies) components may reveal drivers of changes in spectral activity. First, I will present normative maps of complete band power, periodic band power, and the aperiodic exponent obtained from intracranial EEG (iEEG) recordings from 234 participants. Subsequently, I will show abnormalities in spared and resected regions using recordings obtained from focal epilepsy patients. Lastly, I will discuss a the analysis of resting-state magnetoencephalography (MEG) recordings.