Biology subjects

Chiarella, L.

Publications and source records attributed to Chiarella, L..

4 recordsLinked to original sources

bioRxiv2026

Frequency modulations of cortical synchronization in human cortex during wakefulness and sleep

Vigilance states are associated with reproducible reconfigurations of large-scale brain dynamics, reflected in changes in inter-areal phase synchronization and cross-frequency phase-amplitude coupling. Here, we characterized how these two forms of phase-based coordination jointly organize across sleep and wakefulness in the human brain at local and regional levels. We analysed phase-locking value (PLV) and phase-amplitude coupling (PAC) from intracranial stereo-electroencephalography (SEEG) recordings in 46 individuals with drug-resistant focal epilepsy, focusing on contacts outside the epileptogenic zone (non-epileptogenic zone, nEZ) to define physiological coupling profiles and comparing them with contacts within the EZ. For each subject, representative epochs of wakefulness, NREM sleep stages N2 and N3, and REM sleep were examined. Across vigilance states, large-scale phase synchronization exhibited distinct spectral fingerprints. Theta and sigma synchronization predominated during NREM sleep, beta synchronization increased during REM sleep, and theta interactions characterized wakefulness. PAC showed complementary state-dependent reorganizations: N3 was characterized by delta-driven modulation of broadband high-frequency activity; N2 additionally exhibited theta- and spindle-phase modulation of beta-gamma amplitudes; REM sleep showed reduced coupling; and wakefulness was marked by theta-to-beta interactions. Within vigilance states, epileptogenic regions displayed increased delta and gamma synchronization and enhanced delta-to-beta/gamma PAC, most prominently during N2 sleep and wakefulness, whereas differences between EZ and nEZ tissue were attenuated during REM sleep. Using partial least squares analysis, we further identified system-specific patterns of PLV-PAC covariation, with prominent involvement of temporal networks during NREM sleep and visual and limbic systems during REM sleep. Together, these findings delineate a frequency-specific, state-dependent architecture linking phase synchronization and phase-amplitude coupling in the human brain and describe how epileptogenic networks deviate from physiological coupling profiles across vigilance states.

neuroscience↗

bioRxiv2026

REM sleep reconfigures large-scale network dynamics: a link to its suppressive role in epilepsy

Converging evidence suggests that human brain activity operates near a critical-like regime in which balanced excitation and inhibition support efficient large-scale communication. The brains proximity to criticality may be dynamically reset across the sleep-wake cycle and altered by epilepsy, leading to aberrant oscillatory dynamics. Building on recent work demonstrating a tripartite interaction between networks synchronization, oscillatory amplitude bistability, and cross-frequency coupling in the human brain that seems to favour epileptic activity, we examined how this interaction, and its underlying large-scale dynamics are modulated across vigilance states. We analyzed overnight recordings from 20 patients with drug-resistant epilepsy undergoing presurgical evaluation and selected overall 20 minutes of continuous, artifact free stereo-electroencephalography (SEEG) spanning REM sleep, NREM stages N2 and N3, and eyes-closed resting wakefulness. Across states, we quantified phase synchronization, phase-amplitude coupling, bistability and their correlation. REM sleep was consistently associated with a reduction of these dynamics relative to NREM sleep and wakefulness. Importantly, the canonical correlation between these measures -- reflecting the strength of the tripartite interaction -- was significantly weaker during REM sleep. These findings indicate that vigilance states modulate this previously identified multiscale interaction in human brain networks and suggest that the reduced epileptogenicity of REM sleep can be associated with a disruption of coordinated synchronization, coupling, and bistable dynamics at the large-scale network level.

neuroscience↗

bioRxiv2025

Sleep-Modulated Cross-Frequency Coupling Between δ Phase and β-γ Bistability: A System-Level Modulation of Epileptic Activity

ObjectiveWhile slow waves in {delta} (0.5-4 Hz) characterize NREM sleep, in patients with sleep-related epilepsy, seizures most frequently emerge during NREM stage 2, known to be promoted by {delta}-band instability. Meanwhile, the epileptogenic zone (EZ) shows localized bistability in {beta}-{gamma} band (15-200 Hz) neuronal oscillations--indicating a catastrophic shift toward seizure. We aim to clarify the mechanistic link between {delta}-band synchrony and {beta}-{gamma} band bistability in epilepsy. MethodsWe studied a cohort of fourteen patients with Sleep Hypermotor Epilepsy (22.3 {+/-} 10.8 years old; 7 males). 7-9-hour stereo-EEG sleep recordings were segmented into 10-minute of uninterrupted, interictal N2 and N3 epochs, and phase synchrony, phase-amplitude coupling (PAC), and bistability were assessed. Canonical correlation was examined to answer whether PAC links {delta}-phase to {beta}-{gamma} bistability. ResultsCompared to non-EZ, the EZ exhibited larger 15-200 Hz bistability along with stronger 2-8 Hz and 15-100 Hz synchrony throughout N2 and N3. Compared to N3, N2 showed stronger PAC between 2-30 Hz phases in the non-EZ and 5-150 Hz amplitudes in the EZ. Canonical correlations between {delta}-phase modulated PAC and both bistability and synchrony were identified during N2 (r = 0.86 and 0.82) and N3 (r = 0.84 and 0.80), with the strongest contributors being 2-4 Hz synchrony and bistability in 2-4 Hz and 15-200 Hz bands. Correlations between interictal spikes and canonical covariates of bistability and PAC (r2 = 0.62 for N2 and 0.56 for N3) validated their relevance to epileptogenicity. Significance{delta}-band synchrony and {beta}-{gamma} band bistability are not isolated epileptogenic mechanisms but likely act synergistically, playing a pivotal role in seizure generation through the coupling of {delta} phases and {beta}-{gamma} amplitudes across large networks, with significant contributions from non-epileptogenic tissues. Key pointsO_LIStrong {beta}-{gamma} bistability in neuronal oscillations localizes the EZ throughout N2 and N3 sleep. C_LIO_LIElevated {delta}-band phase synchrony characterizes the EZ and its functional neighbors throughout N2 and N3 sleep. C_LIO_LIa-band synchrony modulates local {beta}-{gamma} bistability through PAC, with significant contributions from non-EZ tissues. C_LI

neuroscience↗

bioRxiv2022

NLP-based tools for localization of the Epileptogenic Zone in patients with drug-resistant focal epilepsy

BackgroundDrug-resistant focal epilepsy, defined by failure of two antiepileptic drugs, affects about 30% of patients with epilepsy. Epilepsy surgery may represent an alternative options for this population. However, defining the epileptogenic zone to be surgically removed requires highly specialised medical expertise as well as advanced technologies. The aim of this work is building a cost-effective support system based on text, in particular based on the semiological descriptions of the seizures (temporal vs extratemporal lobe; right vs left hemisphere), in order to predict the localization of seizure origin. MethodsAmong a population of 121 surgically treated and seizure-free drug-resistant patients suffering with focal epilepsy, recruited at the Niguarda Hospital in Milan, we extracted a total number of 509 descriptions of seizures. After a data pre-processing phase, we used natural language processing tools to build numerical representations of the seizures descriptions, both using embedding and countbased methods. We then used machine learning models performing a binary classification into right/left and temporal/extra-temporal. ResultsAll predictive models show a better performance when using the representations relying on embedding models respect to count-based ones. Between all the combinations of representations and classifiers, the best performance obtained in terms of F1-score is 84.7% {+/-} 0.6. DiscussionThis preliminary work reached encouraging results considering both localization tasks. The main advantage is that no specific knowledge about epilepsy is used to build the models, rendering our pipeline applicable also in other scenarios. The major limitation lies in the fact that the text is highly specific to the writer.

bioengineering↗