Rationale:
Previous quantitative analysis regarding intracranial electroencephalography (EEG) studies in epilepsy relied mainly on linear spectral analysis, potentially overlooking the signals' nonlinear dynamics, which are probable keys to understanding the complex integration of neuronal activity in the epileptic brain. Our previous research suggests that
Harmonic pattern (
H pattern)
observed during ictal states could be a potential biomarker for localizing the epileptogenic zone. From a physical perspective, the emergence of
H pattern is likely predominantly driven by nonlinear effects. Therefore, an in-depth analysis of the nonlinear properties harbored by
H pattern is expected to provide new perspectives and methodologies for better elucidating the intricate mechanisms and spatiotemporal propagation patterns of seizures.
Methods: We consecutively included 57 patients with drug-resistant focal epilepsy who underwent stereo-EEG evaluation and resective surgery. A focal onset ictal pattern of all the patients was determined by visual identification. Time-frequency maps were generated using Morlet wavelet transform analysis and further validated by multitaper method. The
H pattern was defined as multiple equidistant high-density frequency bands depicted in the time-frequency map. A threshold of Q3 was employed to confirm channels expressing dominant
H pattern (d
H pattern). Bispectral analysis was used to describe the nonlinear properties of
H pattern, and normalized bicoherence, biphase, skewness, and asymmetry were subsequently acquired using this method. Transfer function modeling was utilized to analyze the propagation between regions expressing d
Hpattern as well as those expressing non
Hpattern and differences in nonlinear effects were evaluated then.
Results: The
H pattern can be accounted by EEG waveshapes, including sharper troughs, sharp extended troughs and peaks, and asymmetric waveshapes. Using nonlinear analysis, it showed that d
H pattern exhibited stronger nonlinear effects than that of non-
dH pattern, and these effects informs a spatiotemporal propagation from regions expressing d
H pattern to those expressing non-dH pattern.
Conclusions: Our findings unveil the influence of nonlinear phenomena on EEG dynamics during seizures, enriching our grasp of the ictal network. The potency of these non-linear interactions orchestrates specific physiological and pathological functions and can serve as a marker for evaluating epileptogenicity.
Funding:
National Natural Science Foundation of China General Program: 82171437; Zhejiang Province Natural Science Foundation Major Project LD24H090003.
复制再试一次分享