18 November 2010 in Epilepsia
Purpose: We assessed 636 epileptic spasms seen in 11 children (median 44 spasms per child) and determined the spatial and temporal characteristics of ictal high‐frequency oscillations (HFOs) in relation to the onset of spasms. Methods: Electrocorticography (ECoG) signals were sampled from 104–148 cortical sites per child, and the dynamic changes of ictal HFOs were animated on each individual’s three‐dimensional (3D) magnetic resonance (MR) image surface. Key Findings: Visual assessment of ictal ECoG recordings revealed that each spasm event was characterized by augmentation of HFOs. Time‐frequency analysis demonstrated that ictal augmentation of HFOs at 80–200 Hz was most prominent and generally preceded those at 210–300 Hz and at 70 Hz and slower. Recruitment of HFOs in the rolandic cortex preceded the clinical onset objectively visualized as electromyographic deflection. The presence or absence of ictal motor symptoms was related more to the amplitude of HFOs in the Rolandic cortex than in the seizure‐onset zone. In a substantial proportion of epileptic spasms, seizure termination began at the seizure‐onset zone and propagated to the surrounding areas; we referred to this observation as the “ictal doughnut phenomenon.” Univariate analysis suggested that complete resection of the sites showing the earliest augmentation of ictal HFOs was associated with a good surgical outcome. Significance: Recruitment of HFOs at 80–200 Hz in the rolandic area may play a role in determining seizure semiology in epileptic spasms. Our study using macroelectrodes demonstrated that ictal HFOs at 80–200 Hz preceded those at 210–300 Hz.
13 September 2011 in Epilepsia
Previous studies of epileptic spasms reported that ictal events were associated with high‐frequency oscillations (HFOs) or delta waves involving widespread regions. We determined whether ictal HFOs at 80–200 Hz were coupled with a phase of slow‐wave, whether ictal slow‐waves were diffusely or locally synchronous signals, and whether the mode of coupling between HFOs and slow‐wave phases differed between ictal and interictal states. We studied 11 children who underwent extraoperative electrocorticography (ECoG) recording. The phases and amplitudes of slow‐waves were measured at the peak of ictal and interictal HFOs in the seizure‐onset sites. Ictal HFOs were locked tightly to the phase of slow‐wave at ≤1 Hz. Ictal slow‐waves propagated from the seizure‐onset site to other regions. In contrast, interictal HFOs in the seizure‐onset site were loosely locked to the phase of slow‐wave at ≤1 Hz but tightly to that of ≥3‐Hz. Ictal slow‐waves coupled with HFOs can be explained as near‐field and locally synchronized potentials generated by the neocortex rather than far‐field potentials generated by subcortical structures. Ictal slow‐waves in epileptic spasms may be generated by a mechanism different from what generates interictal HFOs–slow‐wave complexes.
1 July 2018 in Clinical Neurophysiology
3 November 2021 in Brain Communications
Intracranially recorded interictal high-frequency oscillations have been proposed as a promising spatial biomarker of the epileptogenic zone. However, its visual verification is time-consuming and exhibits poor inter-rater reliability. Furthermore, no method is currently available to distinguish high-frequency oscillations generated from the epileptogenic zone (epileptogenic high-frequency oscillations) from those generated from other areas (non-epileptogenic high-frequency oscillations). To address these issues, we constructed a deep learning-based algorithm using chronic intracranial EEG data via subdural grids from 19 children with medication-resistant neocortical epilepsy to: (i) replicate human expert annotation of artefacts and high-frequency oscillations with or without spikes, and (ii) discover epileptogenic high-frequency oscillations by designing a novel weakly supervised model. The ‘purification power’ of deep learning is then used to automatically relabel the high-frequency oscillations to distill epileptogenic high-frequency oscillations. Using 12 958 annotated high-frequency oscillation events from 19 patients, the model achieved 96.3% accuracy on artefact detection (F1 score = 96.8%) and 86.5% accuracy on classifying high-frequency oscillations with or without spikes (F1 score = 80.8%) using patient-wise cross-validation. Based on the algorithm trained from 84 602 high-frequency oscillation events from nine patients who achieved seizure-freedom after resection, the majority of such discovered epileptogenic high-frequency oscillations were found to be ones with spikes (78.6%, P < 0.001). While the resection ratio of detected high-frequency oscillations (number of resected events/number of detected events) did not correlate significantly with post-operative seizure freedom (the area under the curve = 0.76, P = 0.06), the resection ratio of epileptogenic high-frequency oscillations positively correlated with post-operative seizure freedom (the area under the curve = 0.87, P = 0.01). We discovered that epileptogenic high-frequency oscillations had a higher signal intensity associated with ripple (80–250 Hz) and fast ripple (250–500 Hz) bands at the high-frequency oscillation onset and with a lower frequency band throughout the event time window (the inverted T-shaped), compared to non-epileptogenic high-frequency oscillations. We then designed perturbations on the input of the trained model for non-epileptogenic high-frequency oscillations to determine the model’s decision-making logic. The model confidence significantly increased towards epileptogenic high-frequency oscillations by the artificial introduction of the inverted T-shaped signal template (mean probability increase: 0.285, P < 0.001), and by the artificial insertion of spike-like signals into the time domain (mean probability increase: 0.452, P < 0.001). With this deep learning-based framework, we reliably replicated high-frequency oscillation classification tasks by human experts. Using a reverse engineering technique, we distinguished epileptogenic high-frequency oscillations from others and identified its salient features that aligned with current knowledge.
3 September 2020 in Clinical Neurophysiology
11 April 2017 in Epilepsia
AGAristea S. Galanopoulou
+3
Summary: Objective: We investigated temporal and spatial characteristics of ictal gamma and beta activity on scalp EEG during spasms in patients with West syndrome (WS) to evaluate potential focal cortical onset.Methods: A total of 1,033 spasms from 34 patients with WS of various etiologies were analyzed on video‐electroencephalography (EEG) using time‐frequency analysis. Ictal gamma (35–90 Hz) and beta (15–30 Hz) activities were correlated with visual symmetry of spasms, objective EMG (electromyography) analysis, and etiology of WS.Results: Prior to the ictal motor manifestation, focal ictal gamma activity emerged from one hemisphere (71%, 24/34) or from midline (26%, 9/34), and was rarely simultaneously bilateral (3%, 1/34). Focal ictal beta activity emerged from either one hemisphere (68%, 23/34) or from midline (32%, 11/34). Onsets of focal ictal gamma and beta activity were most commonly observed around the parietal areas. Focal ictal gamma activity propagated faster than ictal beta activity to adjacent electrodes (median: 65 vs. 170 msec, p < 0.01), and to contralateral hemisphere (median: 100 vs. 170 msec, p = 0.01). Asymmetric peak amplitude of ictal gamma activity in the centroparietal areas (C3‐P3 vs. C4‐P4) correlated with asymmetric semiology. On the other hand, most of the visually symmetric spasms showed asymmetry in peak amplitude and interhemispheric onset latency difference in both ictal gamma and beta activity.Significance: Spasms may be a seizure with focal electrographic onset regardless of visual symmetry. Asymmetric involvement of ictal gamma activity to the centroparietal areas may determine the motor manifestations in WS. Scalp EEG ictal gamma and beta activity may be useful to demonstrate localized seizure onset in infants with WS.