Interrater reliability of interictal EEG waveforms in Lennox-Gastaut Syndrome.

OBJECTIVE: Identification of EEG waveforms is critical for diagnosing Lennox-Gastaut Syndrome (LGS) but is complicated by the progressive nature of the disease. Here, we assess the interrater reliability (IRR) among pediatric epileptologists for classifying EEG waveforms associated with LGS. METHODS: A novel automated algorithm was used to objectively identify epochs of EEG with transient high power, which were termed events of interest (EOIs). The algorithm was applied to EEG from 20 LGS subjects and 20 healthy controls during NREM sleep, and 1350 EOIs were identified. Three raters independently reviewed the EOIs within isolated 15-second EEG segments in a randomized, blinded fashion. For each EOI, the raters assigned a waveform label (spike and slow wave, generalized paroxysmal fast activity, seizure, spindle, vertex, muscle, artifact, nothing, or other) and indicated the perceived subject type (LGS or control). RESULTS: Labeling of subject type had 85% accuracy across all EOIs and an IRR of κ =0.790, suggesting that brief segments of EEG containing high-power waveforms can be reliably classified as pathological or normal. Waveform labels were less consistent, with κ =0.558, and the results were highly variable for different categories of waveforms. Label mismatches typically occurred when one reviewer selected "nothing," suggesting that reviewers had different thresholds for applying named labels. SIGNIFICANCE: Classification of EEG waveforms associated with LGS has weak IRR, due in part to varying thresholds applied during visual review. Computational methods to objectively define EEG biomarkers of LGS may improve IRR and aid clinical decision-making.

Evolution of Cortical Functional Networks in Healthy Infants.

During normal childhood development, functional brain networks evolve over time in parallel with changes in neuronal oscillations. Previous studies have demonstrated differences in network topology with age, particularly in neonates and in cohorts spanning from birth to early adulthood. Here, we evaluate the developmental changes in EEG functional connectivity with a specific focus on the first 2 years of life. Functional connectivity networks (FCNs) were calculated from the EEGs of 240 healthy infants aged 0-2 years during wakefulness and sleep using a cross-correlation-based measure and the weighted phase lag index. Topological features were assessed via network strength, global clustering coefficient, characteristic path length, and small world measures. We found that cross-correlation FCNs maintained a consistent small-world structure, and the connection strengths increased after the first 3 months of infancy. The strongest connections in these networks were consistently located in the frontal and occipital regions across age groups. In the delta and theta bands, weighted phase lag index networks decreased in strength after the first 3 months in both wakefulness and sleep, and a similar result was found in the alpha and beta bands during wakefulness. However, in the alpha band during sleep, FCNs exhibited a significant increase in strength with age, particularly in the 21-24 months age group. During this period, a majority of the strongest connections in the networks were located in frontocentral regions, and a qualitatively similar distribution was seen in the beta band during sleep for subjects older than 3 months. Graph theory analysis suggested a small world structure for weighted phase lag index networks, but to a lesser degree than those calculated using cross-correlation. In general, graph theory metrics showed little change over time, with no significant differences between age groups for the clustering coefficient (wakefulness and sleep), characteristics path length (sleep), and small world measure (sleep). These results suggest that infant FCNs evolve during the first 2 years with more significant changes to network strength than features of the network structure. This study quantifies normal brain networks during infant development and can serve as a baseline for future investigations in health and neurological disease.

An Effortful Approach to Social Affiliation in Schizophrenia: Preliminary Evidence of Increased Theta and Alpha Connectivity during a Live Social Interaction.

People with schizophrenia often experience a profound lack of motivation for social affiliation-a facet of negative symptoms that detrimentally impairs functioning. However, the mechanisms underlying social affiliative deficits remain poorly understood, particularly under realistic social contexts. Here, we investigated subjective reports and electroencephalography (EEG) functional connectivity in schizophrenia during a live social interaction. Individuals with schizophrenia (n = 16) and healthy controls (n = 29) completed a face-to-face interaction with a confederate while having EEG recorded. Participants were randomly assigned to either a Closeness condition designed to elicit feelings of closeness through self-disclosure or a Small-Talk condition with minimal disclosure. Compared to controls, patients reported lower positive emotional experiences and feelings of closeness across conditions, but they showed comparably greater subjective affiliative responses for the Closeness (vs. Small-Talk) condition. Additionally, patients in the Closeness (vs. Small-Talk) condition displayed a global increase in connectivity in theta and alpha frequency bands that was not observed for controls. Importantly, greater theta and alpha connectivity was associated with greater subjective affiliative responding, greater negative symptoms, and lower disorganized symptoms in patients. Collectively, findings indicate that patients, because of pronounced negative symptoms, utilized a less efficient, top-down mediated strategy to process social affiliation.

Computational characteristics of interictal EEG as objective markers of epileptic spasms.

OBJECTIVE: Favorable neurodevelopmental outcomes in epileptic spasms (ES) are tied to early diagnosis and prompt treatment, but uncertainty in the identification of the disease can delay this process. Therefore, we investigated five categories of computational electroencephalographic (EEG) measures as markers of ES. METHODS: We measured 1) amplitude, 2) power spectra, 3) Shannon entropy and permutation entropy, 4) long-range temporal correlations, via detrended fluctuation analysis (DFA) and 5) functional connectivity using cross-correlation and phase lag index (PLI). EEG data were analyzed from ES patients (n = 40 patients) and healthy controls (n = 20 subjects), with multiple blinded measurements during wakefulness and sleep for each patient. RESULTS: In ES patients, EEG amplitude was significantly higher in all electrodes when compared to controls. Shannon and permutation entropy were lower in ES patients than control subjects. The DFA intercept values in ES patients were significantly higher than control subjects, while DFA exponent values were not significantly different between the groups. EEG functional connectivity networks in ES patients were significantly stronger than controls when based on both cross-correlation and PLI. Significance for all statistical tests was p < 0.05, adjusted for multiple comparisons using the Benjamini-Hochberg procedure as appropriate. Finally, using logistic regression, a multi-attribute classifier was derived that accurately distinguished cases from controls (area under curve of 0.96). CONCLUSIONS: Computational EEG features successfully distinguish ES patients from controls in a large, blinded study. SIGNIFICANCE: These objective EEG markers, in combination with other clinical factors, may speed the diagnosis and treatment of the disease, thereby improving long-term outcomes.

Schizotypy dimensions are associated with altered resting state alpha connectivity.

The disconnection hypothesis of schizophrenia says that symptoms are explained by dysfunctional connections across a wide range of brain networks. Despite some support for this hypothesis, there have been mixed findings. One reason for these may be the multidimensional nature of schizophrenia symptoms. In order to clarify the relationship between symptoms and brain networks, the current study included individuals at risk for schizophrenia-spectrum disorders who either report extreme levels of positive schizotypy traits (perceptual aberrations and magical ideation, or "PerMag"; n = 23), or an extreme negative schizotypy trait (social anhedonia, or "SocAnh"; n = 19), as well as a control group (n = 18). Resting-state alpha electroencephalography was collected, and functional networks for each subject were measured using the phase-lag index to calculate the connectivity between channel pairs based on the symmetry of instantaneous phase differences over time. Furthermore, graph theory measures were introduced to identify network features exclusive to schizotypy groups. We found that the PerMag group exhibited a smaller difference in node strength and clustering coefficient in frontal/occipital and central/occipital regional comparisons compared to controls, suggesting a more widespread network. The SocAnh group exhibited a larger difference in degree in the central/occipital regional comparison relative to controls, suggesting a localized occipital focus in the connectivity network. Regional differences in functional connectivity suggest that different schizotypy dimensions are manifested at the network level by different forms of disconnections. Taken together, these findings lend further support to the disconnection hypothesis and suggest that altered connectivity networks may serve as a potential biomarker for schizophrenia risk.

Effect of interictal epileptiform discharges on EEG-based functional connectivity networks.

OBJECTIVE: Functional connectivity networks (FCNs) based on interictal electroencephalography (EEG) can identify pathological brain networks associated with epilepsy. FCNs are altered by interictal epileptiform discharges (IEDs), but it is unknown whether this is due to the morphology of the IED or the underlying pathological activity. Therefore, we characterized the impact of IEDs on the FCN through simulations and EEG analysis. METHODS: We introduced simulated IEDs to sleep EEG recordings of eight healthy controls and analyzed the effect of IED amplitude and rate on the FCN. We then generated FCNs based on epochs with and without IEDs and compared them to the analogous FCNs from eight subjects with infantile spasms (IS), based on 1340 visually marked IEDs. Differences in network structure and strength were assessed. RESULTS: IEDs in IS subjects caused increased connectivity strength but no change in network structure. In controls, simulated IEDs with physiological amplitudes and rates did not alter network strength or structure. CONCLUSIONS: Increases in connectivity strength in IS subjects are not artifacts caused by the interictal spike waveform and may be related to the underlying pathophysiology of IS. SIGNIFICANCE: Dynamic changes in EEG-based FCNs during IEDs may be valuable for identification of pathological networks associated with epilepsy.