SGM: a novel time-frequency algorithm based on unsupervised learning improves high-frequency oscillation detection in epilepsy.

OBJECTIVE: We propose a novel automated method called the S-Transform Gaussian Mixture detection algorithm (SGM) to detect high-frequency oscillations (HFO) combining the strengths of different families of previously published detectors. APPROACH: This algorithm does not depend on parameter tuning on a subject (or database) basis, uses time-frequency characteristics, and relies on non-supervised classification to determine if the events standing out from the baseline activity are HFO or not. SGM consists of three steps: the first stage computes the signal baseline using the entropy of the autocorrelation; the second uses the S-Transform to obtain several time-frequency features (area, entropy, and time and frequency widths); and in the third stage Gaussian mixture models cluster time-frequency features to decide if events correspond to HFO-like activity. To validate the SGM algorithm we tested its performance in simulated and real environments. MAIN RESULTS: We assessed the algorithm on a publicly available simulated stereoelectroencephalographic (SEEG) database with varying signal-to-noise ratios (SNR), obtaining very good results for medium and high SNR signals. We further tested the SGM algorithm on real signals from patients with focal epilepsy, in which HFO detection was performed visually by experts, yielding a high agreement between experts and SGM. SIGNIFICANCE: The SGM algorithm displayed proper performance in simulated and real environments and therefore can be used for non-supervised detection of HFO. This non-supervised algorithm does not require previous labelling by experts or parameter adjustment depending on the subject or database considered. SGM is not a computationally intensive algorithm, making it suitable to detect and characterize HFO in long-term SEEG recordings.

Physiological Ripples Associated with Sleep Spindles Differ in Waveform Morphology from Epileptic Ripples.

High frequency oscillations (HFOs, 80-500[Formula: see text]Hz) serve as novel electroencephalography (EEG) markers of epileptic tissue. The differentiation of physiological and epileptic HFO is an important challenge and is complicated by the fact that both types are generated in mesiotemporal structures. This study aimed to identify oscillation features that serve to distinguish physiological ripples associated with sleep spindles and epileptic ripples. We studied 19 patients with chronic intracranial EEG(iEEG) with mesiotemporal implantation and simultaneous scalp EEG. Sleep spindles, ripples and spikes were visually marked during nonrapid eye movement sleep stage 2. Ripples co-occurring with spikes and in seizure onset zone (SOZ) channels but outside of spindles were considered epileptic. The SOZ is defined by the origin of clinical seizures in iEEG. Ripples co-occurring with spindles were considered as models for physiological ripples. A correlation analysis showed a significant ripple amplitude peak - spindle trough - coupling, thus proving their physiological linkage. Epileptic ripples showed significantly higher values in all amplitude features than spindle ripples. All amplitude features and peaks per sample length showed a predictive value for the classification between model physiological ripples and epileptic ripples but indicate that the specificity is not sufficient for a reliable discrimination of single ripple events. The presented results suggest that a secure identification of epileptic ripples may be available to help identify the epileptic focus in the future.