Clinical Neurophysiology

Accurate detection of interictal epileptiform discharges (IEDs) in electroencephalography (EEG) plays a crucial role in epilepsy diagnosis. Our work investigates the classification of IEDs using Artificial Neural Networks (ANNs) trained on EEG data represented in both signal and source space. Source waveforms were computed using equivalent current dipole models fitted using either a 1-parameter fixed-orientation or a 3-parameter projection approach, both localized to a single best-fit position during the rising flank of the IED. The ANN was trained on raw and feature-extracted versions of signal space and source space data. Feature extraction significantly improved performance across all domains. The highest accuracy (0.98) was achieved in signal space using Katz Fractional Dimension (KFD). In source space analyses, the 1-parameter and 3-parameter models achieved a maximum accuracy of 0.84, with statistical features performing best for the fixed-orientation model and KFD for the free orientation model. Additionally, annotations from three independent expert markers showed considerable variability, with ANN performance falling within the range of inter-expert agreement. These findings support the potential of ANN-based tools to assist expert evaluation in future clinical workflows.

More than 15 million patients worldwide suffer from drug-resistant epilepsy (DRE). Surgical removal of the seizure onset zone (SOZ)--the brain region(s) from where seizures arise--is the best available treatment for these patients, with post-surgical outcomes depending on the successful identification and resection of the SOZ. Most commonly, SOZ mapping localizes ictal activity occurring spontaneously or evoked by cortical stimulation (CS) during presurgical evaluation of patients with epilepsy using stereoelectroncephalography (SEEG). Mapping events such as interictal epileptiform discharges (IEDs), paroxysmal hypersynchronic electrical discharges that often occur outside ictal discharges or during CS, have been less used for SOZ localization. We test the hypothesis that IEDs triggered by CS via SEEG investigation can contribute to the mapping of the SOZ. We evaluated the impact of CS on IEDs in patients investigated with SEEG on epilepsy surgery investigation. We recorded intracranial signals from 30 DRE patients (seizure-free post-surgery). Bipolar and high frequency (50 Hz) CS was performed with a pulse width of 300 {micro}s and current spanning 1-8 mA. IEDs were automatically detected pre- and post- stimulation, and their normalized absolute changes were quantified within and outside the SOZ (identified by ictal discharges). We found that IED rates significantly increased post-stimulation compared to pre-stimulation within the SOZ, while no significant change was observed outside the SOZ (Linear mixed effect model, p-value <0.001, and AUC=98% for SOZ and 71% for non-SOZ). This effect was present regardless of whether the stimulation was applied to the SOZ or non-SOZ regions, indicating a broader effect of stimulation on the SOZ. Our results offer a quantitative tool for identifying epileptogenic areas in patients with DRE, enhancing the mapping and localization of the SOZ and potentially improving surgical outcomes.

The anterior nucleus of the thalamus (ANT) mediates cortical-subcortical interactions between the limbic system and is hypothesized to facilitate the early organization of temporal lobe seizures. We set out to investigate the dynamic changes in synchronization parameters between the seizure onset zone (SOZ) and ANT during seizure stages (pre-onset to post-termination) in seven patients (n=26 seizures) with drug-resistant nonlesional temporal lobe epilepsy. Using local field potentials recorded directly from the limbic system and the ANT during stereoelectroencephalography, we confirm that the onset of mesial temporal lobe seizure is associated with increased thalamocortical network excitability and phase-amplitude coupling. The increase in thalamocortical phase synchronization preceded seizure onset, thereby suggesting that the early organization of temporal lobe seizures involves the integration of the ANT within the epileptic network. Towards seizure termination, there is a significant decrease in thalamic excitability, thalamocortical synchronization, and decoupling, thereby suggesting a breakdown in thalamocortical connectivity. A higher disease burden is significantly correlated with increased synchronization between the ANT and epileptic networks. Collectively, the results elucidate mechanistic insights and provide the temporal architecture of thalamocortical interactions that can be targeted in the rational designing of closed-loop seizure abortive interventions. HighlightsO_LIAnterior nucleus of thalamus is coactivated at the onset of temporal lobe seizures C_LIO_LIIncrease thalamocortical synchronization and excitability is observed at seizure onset C_LIO_LISeizure termination is characterized by a breakdown in thalamocortical connectivity C_LIO_LIIncreased seizure burden affects thalamocortical synchronization C_LI

The thalamic nuclei play diverse roles in the initiation, propagation, and termination of temporal lobe seizures. The role of the anterior nucleus of the thalamus (ANT) - a node that is integral to the limbic network is unclear. The objective of this study was to characterize temporal and - spectral patterns of ANT ictal recruitment in drug-resistant temporal lobe epilepsy (TLE). We hypothesized that seizures localized to the temporolimbic network are likely to recruit ANT, and the odds of recruitment were higher in seizures that had altered consciousness. Ten patients undergoing stereo-electroencephalography (SEEG) were recruited prospectively to record field potentials from the ANT. Using epileptogenicity index and line length, we computed the number of seizures that recruited the ANT (recruitment ratio), the recruitment latencies between the ANT and the epileptogenic zone (EZ), and latency of ANT recruitment to clinical manifestation for seventy-nine seizures. We observed that seizures localized to mesial temporal subregions (hippocampus, amygdala, anterior cingulate) have a higher predilection for ANT recruitment, and the recruitment was faster (ranged 5-12 secs) and preceded clinical onset for seizures that impaired consciousness. Seizures that recruited ANT lasted significantly longer (t=1.795, p=0.005). Recruitment latency was inversely correlated to seizure duration (r=-0.78, p=0.004). Electrical stimulation of the EZ induced seizures, in which early recruitment of ANT was confirmed. Stimulation of ANT did not induce a seizure. Finally, we tested the hypothesis that spectral and entropy-based features extracted from thalamic field potentials can distinguish its state of ictal recruitment from other interictal states (including awake, sleep). For this, we employed classification machine learning that discriminated thalamic ictal state from other interictal states with high accuracy (92.8%) and precision (93.1%). Among the features, the emergence of the theta rhythm (4-8 Hz) maximally discriminated the endogenous ictal state from other interictal states of vigilance. These results prompt a mechanistic role for the ANT in the early organization and sustaining of seizures, and the possibility to serve as a target for therapeutic closed-loop stimulation in TLE.

Epilepsy is characterized by recurrent seizures and may also have negative influence on cognitive function. In addition to ictal activity, the epileptic brain also gives rise to interictal epileptiform discharges (IEDs). These IEDs constitute the diagnostic hallmark of epilepsy, and have been linked to impaired memory formation and negative effects on neurodevelopment. The neurophysiological dynamics underlying IED generation seem to resemble those underlying seizure development. Understanding the neurophysiological characteristics surrounding and preceding IED development would hence provide valuable insights into the pathophysiology of the epileptic brain. In order to improve this understanding, we aimed to characterize the dynamical activity changes that occurs immediately prior to an IED onset. We used magnetoencephalography (MEG) recordings from nine focal epilepsy patients to characterize the oscillatory activity preceding IED onsets. Our results showed a systematic and gradual increase in oscillatory delta and theta band activity (1-4 Hz and 4-8 Hz, respectively) during this pre-IED interval, reaching a maximum power at IED onset. These results indicate that the pre-IED brain state is characterized by a gradual synchronization that culminates in the neuronal hypersynchronization underlying IEDs. We discuss how IED generation might resemble seizure development, where physiological brain activity similarly undergoes a gradual synchronization that terminates in seizure onset.

ObjectiveHippocampal ripples are biomarkers of epileptogenicity in patients with epilepsy, and physiological features characterize memory function in healthy individuals. Discriminating between pathological and physiological ripples is important for identifying the epileptogenic (EP) zone; however, distinguishing them from waveforms is difficult. This study hypothesized that the nocturnal synchronization of hippocampal ripples and cortical delta power classifies EP and physiological hippocampi. MethodsWe enrolled 38 patients with electrodes implanted in the hippocampus or the parahippocampal gyrus between April 2014 and March 2023 at our institution. We classified 11 patients (11 hippocampi) into the EP group, who were pathologically diagnosed with hippocampal sclerosis, and five patients (six hippocampi) into the non-epileptogenic (NE) group, whose hippocampi had no epileptogenicity. Hippocampal ripples were detected using intracranial electroencephalography of the hippocampal or parahippocampal electrodes and presented as ripple rates per second. Cortical delta power (0.5-4 Hz) was assessed using cortical electrodes. The Pearson correlation coefficient between the ripple rates and the cortical delta power (CRD) was calculated for the intracranial electroencephalographic signals obtained every night during the recordings. ResultsHippocampal ripples detected from continuous recording for approximately 10 days demonstrated similar frequency characteristics between the EP and NE groups. However, CRDs in the EP group (mean [standard deviation]: 0.20 [0.049]) were significantly lower than those in the NE group (0.67 [0.070], F (1,124) = 29.6, p < 0.0001 (group), F (9,124) = 1.0, p = 0.43 (day); two-way analysis of variance). Based on the minimum CRDs during the 10-day recordings, the two groups were classified with 94.1% accuracy. ConclusionCRD is a biomarker of hippocampal epileptogenicity. Key PointsThe correlation between hippocampal ripple rate and cortical delta power was evaluated for approximately 10 days in patients with drug-resistant epilepsy. Mean correlation coefficients were significantly lower in the epileptogenic group than in the non-epileptogenic group. The minimum value of the correlation coefficients predicts hippocampal sclerosis.

ObjectiveThe study was aimed at developing an automatic system, based on complex network analysis and machine learning, to identify interictal network-based biomarkers in patients with drug-resistant focal epilepsy and no visible anatomical lesions candidate for surgery, able to support the localization of the Epileptogenic Zone (EZ) and to further disclose properties of the interictal epileptogenic network. Methods3 min of interictal SEEG signals, recorded in 18 patients with drug-resistant epilepsy, different EZ localization, negative MRI, were analysed. Patients were divided into seizure-free (SF) and non-seizure free (NSF) groups, according to their post-surgical outcome. After a first step of effective connectivity estimation, hubs were defined through the combination of nine graph theory-based indices of centrality. The values of centrality indices related to these hubs were used as input of an ensemble subspace discriminant classifier. ResultsThe proposed procedure was able to automatically localise the EZ with 98% sensitivity and 59% specificity for SF patients. Moreover, our results showed a clear difference between SF and NSF patients, mainly in terms of false positive rate (i.e., the percentage of NEZ leads classified as EZ), which resulted significantly higher in NSF patients. Lastly, the centrality indexes confirmed a different role of the Propagation Zone in NSF and SF groups. SignificanceResults pointed out that network centrality plays a key role in interictal epileptogenic network, even in case of the absence of anatomical alterations and SEEG epileptic abnormalities, and that the combination of connectivity, graph theory, and machine learning analysis can efficiently support interictal EZ localization. These findings also suggest that poorer post-surgical prognosis can be associated with larger connectivity alteration, with wider "hubs", and with a different involvement of the PZ, thus making this approach a promising biomarker for surgical outcome. Impact statementThe correct localization of the epileptogenic zone is still an unsolved question, mainly based on visual and subjective analysis of electrophysiological recordings, and highly time-consuming due to the needing of ictal recording. This issue is even more critical in patients with negative MRI and extra-temporal EZ localization. The approach proposed in this study represents an innovative and effective tool to reveal interictal epileptogenic network abnormalities, able to support and improve the EZ presurgical identification and to capture differences between poor and good post-surgical outcome

Mesial temporal lobe epilepsy (mTLE) is a common localization of drug-resistant epilepsy in adults. Patients often undergo intracranial EEG (iEEG) monitoring to confirm localization and determine candidacy for focal ablation or resection. Clinicians primarily base surgical decision-making on seizure onset patterns, with imaging abnormalities and information from interictal epileptiform discharge (spikes) used as ancillary data. How the morphology and timing of spikes within multi-electrode sequences may inform surgical planning is unknown, in part due to the lack of measurement methods for large datasets. We hypothesized that patients with mTLE have a distinct mesial-to-lateral spike pattern that differentiates them from other epilepsy localizations. In a multicenter study at the University of Pennsylvania and the Medical University of South Carolina, we analyzed the timing and morphology of spikes and seizure high frequency energy ratio (HFER) in 75 patients with drug-resistant epilepsy. We compared these features across patients with mTLE, temporal neocortical epilepsy, and other localizations. A logistic regression model combining all features predicted a clinical localization of mTLE in unseen patients with an AUC of 0.82 (compared to an AUC of 0.70 for seizure-only features, DeLongs test p = 0.08). Spike rate was the most important feature in the combined model. Modeled probability of mTLE was similar between patients who had a good versus a poor 12-month outcome after undergoing a mesial temporal resection or ablation (p = 0.34). These findings support quantitative spike analysis to supplement analysis of seizures for use in surgical planning.

ObjectiveLocalization of the epileptogenic zone (EZ) requires further refinement. We identified a unique ictal spectral structure, the harmonic pattern (H pattern), which potentially serves as a novel biomarker for localizing the EZ. This study aimed to analyze the clinical significance of the H pattern and to explore its underlying waveform features. MethodsSeventy patients with drug-resistant focal epilepsy, undergoing stereo-EEG (SEEG) evaluation and surgery, were included. Time-frequency maps (TFM) were generated using Morlet wavelet transform analysis. The H pattern was defined as multiple equidistant, high-density bands with varying frequencies on TFM. The upper quartile was employed to confirm contacts expressing dominant H pattern (dH pattern). Bispectral analysis and transfer function modeling were employed to assess nonlinear properties and propagation, respectively. The performance of the dH pattern in evaluating the EZ was compared with other ictal biomarkers. ResultsRegardless of seizure onset patterns, the H pattern commonly occurred during early or late seizure propagation among 57 patients (81.4%). It harbored within specific EEG segments characterized by fast activity and irregular polyspikes. The H pattern often appeared simultaneously across different brain regions at a consistent fundamental frequency, highlighting a crucial stage in seizure propagation characterized by inter-regional synchronization. The dH pattern demonstrated greater nonlinearity compared to the non-dH pattern, as evidenced by bispectral analysis. The waveforms associated with the dH pattern were more stereotyped and showed increased skewness and/or asymmetry. Notably, the complete removal of areas exhibiting the dH pattern, but not high epileptogenicity index ([&ge;]0.3) or seizure onset zone, was independently associated with seizure freedom after surgery. SignificanceThe H pattern provides unique insights into ictal neural dynamics. Additionally, it is a novel and alternative approach for measuring the EZ over an extended ictal time window. KEY POINTSO_LIThe harmonic pattern (H pattern) is commonly present in focal epileptic seizures and can help to improve the accuracy of EZ localization over an extended time window. C_LIO_LIThe H pattern is a spectral signature of waveform skewness or asymmetry. The dominant H pattern reflects a stronger nonlinearity of ictal EEG signals. C_LIO_LIThe H pattern can appear simultaneously in different areas with a consistent fundamental frequency, indicating a key stage of inter-regional synchronization. C_LI

Epileptic seizures and interictal epileptiform discharges (IEDs) are strongly influenced by sleep and circadian rhythms. However, human data on the effect of sleep on neuronal behavior during interictal activity have been lacking. We analyzed EEG data from epileptic patients implanted with macro and micro electrodes targeting mesial temporal structures. Sleep staging was performed on concomitantly recorded polysomnography and video-EEG. Automated IED detection identified thousands of IEDs per patient. Both the rate and amplitude of IEDs were increased with deeper stages of NREM sleep. Single unit activity (SUA) and multi-unit activity (MUA) increased their firing during the IED spike, and strongly decreased during the subsequent slow wave. These time-locked firing rate modulations were shown to increase during deeper stages of NREM sleep. Finally, during resting behaviour, neuronal firing rate, bursting rate and firing regularity were all shown to progressively decrease with deeper stages of NREM sleep.

BackgroundEpilepsy has long been conceptualised as a disorder in which aberrant brain dynamics extend beyond the epileptogenic zone. Evidence demonstrates that loss of entropy is a generic feature of pathological dynamics in the brain, including the ictal state. However, the impact of recurrent seizures on entropy in the interictal state remains unknown. MethodsResting state magnetoencephalography (MEG) scans and resection masks of 32 individuals with epilepsy who had Engel I outcome post-surgery were retrospectively retrieved. Using co-registered FreeSurfer parcellations, we reconstructed the source localised MEG time series and computed sample entropy for 114 regions of interest. We then tested the association of entropy with the resected volume of the brain, and additional clinical variables including the age of seizure onset, seizure frequency and duration of epilepsy. To further understand the temporal relationship between seizure onset and entropy in the interictal state, we collected and computed sample entropy for week-long EEG traces from leucine-rich glioma inactivated 1 monoclonal antibody (LGI1-mAb) rodent models of autoimmune encephalitis (n=5) and control rats (n=5). ResultsIn individuals with epilepsy, a lower age of seizure onset was associated with lower mean sample entropy of the whole cortex (Spearmans rho =0.60, p<0.001; partial correlation =0.41, p=0.021). Entropy did not differ between the resected and non-resected regions of the brain. Furthermore, LGI1-mAb treated rodents showed a persistent decrease in sample entropy as compared to control rats, after the onset of seizures, and this difference was greatest during periods of highest seizure frequency (p<0.001). ConclusionRecurrent seizures are associated with a persistent decrease in entropy, even in the interictal state, and this decrease was found to be most profound and affecting the whole cortex in patients who had a lower age of seizure onset.

ObjectiveSignal analysis biomarkers, in an intra-operative setting, may be complementary tools to guide and tailor the resection in drug-resistant epilepsy patients. Unbiased assessment of biomarker performances are needed to evaluate their clinical usefulness and translation. We defined a realistic ground-truth scenario and compared the effectiveness of different biomarkers alone and combined to localize epileptogenic tissue. MethodsWe investigated the performances of univariate, bivariate and multivariate signal biomarkers applied to 1 minute inter-ictal intra-operative electrocorticography to discriminate between electrodes covering normal or pathologic activity in 47 drug-resistant people with epilepsy (temporal and extra-temporal) who had been seizure-free one year after the operation. ResultsThe best result using a single biomarker was obtained using the phase-amplitude coupling measure for which the epileptogenic tissue was localized in 16 out of 47 patients. Combining the whole set of biomarkers provided an improvement of the performances: 20 out of 47 patients. Repeating the analysis only on the temporal-lobe resections we reached a sensitivity of 93% (28 out of 30) combining all the biomarkers. ConclusionWe suggest that the assessment of biomarker performances on a ground-truth scenario is required to have a proper estimate on how biomarkers translate into clinical use. Phase-amplitude coupling seems the best performing single biomarker and combining biomarkers improves localization of epileptogenic tissue. However, sensitivity achieved is not adequate for the usage as a tool in the operation theater, but it can improve the understanding of pathophysiological process.

ObjectiveTo determine the impact of epilepsy duration and interictal spikes on functional connectivity in children with Self-Limited Epilepsy with Centrotemporal Spikes (SeLECTS). MethodsConnectivity was calculated from electroencephalograms (EEGs) of 68 children with SeLECTS and 65 age and sex-matched controls using the weighted phase lag index. SeLECTS EEGs were categorized by epilepsy duration (shorter or longer than 6 months) to assess progressive connectivity changes. To investigate the impact of spikes on connectivity, 19 SeLECTS patients who underwent two EEGs were analyzed longitudinally, comparing those whose spikes persisted versus resolved over time. Analyses focused on connectivity during sleep. ResultsConnectivity increased with epilepsy duration, being lowest in controls, intermediate in patients with shorter epilepsy duration, and highest in those with longer duration. Changes were initially greatest within the right occipital region and became more widespread with longer epilepsy duration. Longitudinally, patients with persistent spikes showed increasing connectivity over time, while those with spike resolution demonstrated decreasing connectivity, resulting in significant between-group differences. ConclusionsFunctional connectivity in SeLECTS increases progressively with epilepsy duration and spike exposure, suggesting that ongoing spikes drive neural network alterations. SignificanceSpikes are a potential treatment target to prevent progressive brain network disruption and preserve cognitive outcomes.

Objective: Needle electromyography (nEMG) is essential for diagnosing neuromuscular disorders but is invasive and often painful. We employed single-channel bipolar surface EMG (sEMG) analyzed with a novel wavelet-based analytical approach, Detecting and Extracting Elemental Wave Components based on a Wavelet Coefficient Set (DEWCS) and investigated whether fasciculation-related activity could be identified. Methods: In this prospective study, 28 patients undergoing nEMG for suspected neuromuscular disorders and 13 healthy controls were included. Resting-state sEMG was recorded from selected muscles using single-channel bipolar active electrodes at a high sampling rate. DEWCS was used to extract indices reflecting fast- and slow-type motor unit (MU)-related activity. These standardized indices were evaluated against nEMG-detected fasciculation potentials using generalized estimating equation logistic regression to account for within-subject clustering. Diagnostic performance was assessed by receiver operating characteristic analysis. Results: A total of 67 muscles from 38 participants were analyzed. Indices of fast- and slow-type MU-related activity were significantly associated with fasciculation potentials (slow: OR 5.10, p = 0.0041; fast: OR 2.38, p = 0.0162). The combined model showed excellent discrimination (area under the curve = 0.97), outperforming either index alone. Muscle region had no significant effect. Conclusions: A single-channel bipolar sEMG setup combined with DEWCS detected fasciculation-related activity with promising accuracy. This method may serve as a non-invasive surrogate marker of lower motor neuron involvement. Further validation in larger cohorts is warranted. Significance: This non-invasive sEMG approach may help detect fasciculation-related activity and complement nEMG in neuromuscular diagnostics.

Human brain cell assemblies fire electrical impulses up to very high frequencies, which limit is not yet known. Using advanced intracranial microEEG recordings in a cohort of epileptic patients, we newly identified short-lasting oscillations in frequencies between 2 and 8 kHz. These ultra fast oscillations were consistently and locally detected in epileptic hippocampi but were extremely rare in non-epileptic brain tissue. The discovered electrophysiological phenomena thus may reflect neuronal hyperexcitability.

This commentary addresses the debate regarding the predictive value of multiday seizure cycles versus simple statistical baselines. Multidien seizure cyclicity is a prevalent, patient-specific phenomenon with promise for epilepsy management. We challenge the assertion that cycle tracking is no better than a 90-day moving average, which is an inherently retrospective model that lags changes in seizure likelihood. We compared a causal cyclic forecast to a prospectively applied moving average across a large seizure diary cohort (n=768) and two gold-standard chronic EEG cohorts (n=24). At the group level for the EEG and diary cohorts, cycle tracking demonstrated significantly superior accuracy to the moving average for both hourly and daily forecasts (p < 0.0001). These results confirm that event-based cyclical models offer more accurate, simulated real-world forecasts. We conclude that robust forecasting tools must prioritize the detection and modeling of seizure cycles to move beyond simple baseline performance and provide actionable clinical utility.

Background and ObjectivesTimely diagnosis and effective treatment of Lennox-Gastaut Syndrome (LGS) improve prognosis and lower healthcare costs, but the transition from infantile epileptic spasms syndrome (IESS) to LGS is highly variable and insidious. Objective biomarkers are needed to monitor this progression and guide clinical decision making. MethodsWe retrospectively collected longitudinal EEG data at the Childrens Hospital of Orange County from fifteen children who were diagnosed with IESS and later with LGS between 2012 and 2021. EEGs were from IESS and LGS diagnoses, between the two diagnoses, and following LGS diagnosis. Functional connectivity networks were calculated using a cross-correlation-based method and assessed relative to diagnostic timepoint, treatment response, presence of clinical markers of disease, age, and amplitude of interictal spikes. ResultsConnectivity strength was high at LGS diagnosis and decreased after favorable response to treatment, but it remained stable or increased when response was unfavorable. In all subjects, connectivity strength was higher at the time of LGS diagnosis than at the preceding timepoint. Presence of clinical markers of LGS were associated with high connectivity strength, but no single marker predicted connectivity strength. DiscussionComputational EEG analysis can be used to map the evolution from IESS to LGS. Changes in connectivity may enable prediction of impending LGS and treatment response monitoring, thus facilitating earlier LGS treatment and guiding medical management. Key pointsO_LIEEG functional connectivity analysis can track progression from infantile epileptic spasms syndrome (IESS) to Lennox-Gastaut Syndrome (LGS). C_LIO_LIHigh connectivity strength at LGS diagnosis decreases with favorable treatment response but remains high with poor response. C_LIO_LIClinical LGS markers correlate with high connectivity, but no single marker predicts connectivity strength. C_LIO_LIEEG functional connectivity analysis may help predict LGS onset, enabling early intervention and improving prognosis. C_LI

Objective: Computational magnetoencephalography (MEG) interictal epileptiform discharge (IED) detectors have mainly used generalized MEG-only models, whereas clinical MEG interpretation routinely integrates simultaneous electroencephalography (EEG) and includes MEG-unique or MEG-dominant discharges. We developed a patient-specific EEG-MEG IED detector and evaluated event-wise prediction stability across models and the effect of adding EEG to MEG-based prediction. Methods: Seventeen patients undergoing clinical EEG-MEG evaluation for epilepsy were retrospectively analyzed. Clinically accepted dipole-review IEDs were treated as positive events, and nonannotated events were sampled as negatives. Logistic regression (LR), random forest (RF), and a lightweight three-dimensional ResNet were trained separately within each patient using EEG-only, MEG-only, and combined EEG-MEG (EMEG) inputs. Primary performance metrics were the area under the receiver operating characteristic curve (ROC-AUC) and average precision. Event-wise stability was assessed using rank disagreement, rank volatility, and class-aware distribution quotient analysis. Results: Aggregate discrimination was high across models and modalities. Median ROC-AUCs for EEG, MEG, and EMEG were 0.850, 0.890, and 0.880 for LR; 0.880, 0.860, and 0.910 for RF; and 0.920, 0.960, and 0.960 for ResNet. Despite comparable aggregate performance, event-wise analysis revealed model-dependent prediction behavior. ResNet showed significantly lower non-IED rank volatility than classical machine learning models and lower non-IED rank disagreement, particularly compared with RF. Adding EEG to MEG was associated with more favorable class-aware event-wise positioning in most events, while MEG-unique/dominant cases showed greater relative MEG contribution. Conclusions: Patient-specific EEG-MEG IED detection revealed clinically meaningful event-wise differences not captured by aggregate metrics. Simultaneous EEG complemented MEG-based detection, while MEG contribution remained prominent in MEG-dominant cases, supporting multimodal patient-specific IED event prioritization.

ObjectiveStroke-related corticospinal tract (CST) disruption causes lasting hand impairments, but many stroke survivors retain some residual CST connections. In neurotypical adults, motor cortex (M1) TMS interventions can strengthen CST transmission when coupled to EEG brain states reflecting heightened M1 excitability. Because stroke alters the relationship between these brain states and cortical excitability, we aimed to identify poststroke brain states that accurately capture ipsilesional M1 excitability. We hypothesized that heightened ipsilesional M1 excitability would be represented by a common, group-level EEG pattern and a participant- specific, personalized pattern. MethodsWe acquired single-pulse TMS-EEG-EMG datasets in 15 chronic stroke survivors with residual CST connections. We then identified group-level and individual-specific EEG power patterns that distinguished between high and low ipsilesional M1 excitability states. ResultsAt the group level, bilateral sensorimotor mu power was significantly suppressed during high versus low excitability states, but this suppression did not correlate with hand impairment severity or trait-level ipsilesional M1 excitability. At the individual level, spatiotemporally varied EEG activity patterns distinguished between excitability states, but these patterns were only present in 60% of individuals. Conclusion and SignificanceThis study is the first to systematically characterize poststroke EEG brain states reflecting ipsilesional M1 excitability. Findings suggest that individual-specific EEG patterns may inconsistently index ipsilesional M1 excitability and instead identify bilateral sensorimotor mu power suppression as a group-level excitability marker that is present across the full spectrum of poststroke hand impairment. HighlightsO_LIWe analyzed TMS-EEG-EMG to identify group and individual level ipsilesional motor cortical excitability states in chronic stroke C_LIO_LIBilateral sensorimotor mu suppression marked heightened ipsilesional motor cortical excitability across hand impairment severity C_LIO_LI60% participants had individual level scalp patterns linked to motor cortical excitability states, challenging their reliability C_LI

BackgroundElectroencephalography (EEG) can be combined with transcranial magnetic stimulation (TMS) to perform brain-state-dependent stimulation. EEG-TMS studies have shown that corticospinal excitability, as measured via motor evoked potentials (MEPs), is modulated by pre-stimulus periodic EEG features, such as sensorimotor mu-rhythm phase and power. However, the influence of aperiodic brain activity on corticospinal excitability is largely unexplored. ObjectivesWe evaluated the relationship between aperiodic and periodic mu-power, aperiodic exponent, and mu-phase on MEP amplitudes using EEG-TMS. MethodsWe applied 800 single TMS pulses to the left primary motor cortex in 78 healthy adults. We calculated aperiodic/periodic mu-power, aperiodic exponent, and mu-phase for each trial from the pre-stimulus C3-Hjorth transformed EEG. MEP amplitudes were extracted from the right first dorsal interosseous muscle. A linear mixed-effects model assessed relationships between MEP amplitudes and EEG features, with interactions between mu-phase and all other EEG features. ResultsAperiodic and periodic mu-power, aperiodic exponent, and mu-phase significantly modulated MEP amplitudes. Higher aperiodic/periodic mu-power was associated with larger MEP amplitudes, while higher aperiodic exponent was associated with smaller MEP amplitudes. We found a significant interaction effect of aperiodic exponent and mu-phase on MEP amplitude. Aperiodic exponent was negatively associated with MEPs for trough, rising, falling phases, but positively associated with MEPs for peak phase. ConclusionsAperiodic and periodic features of brain activity are reflective of dissociable corticospinal excitability states. Future brain-state-dependent TMS interventions may include aperiodic EEG features, such as aperiodic mu-power and exponent, in addition to the well-established periodic features.