Automated Interpretation of Clinical Electroencephalograms Using Artificial Intelligence.

Importance: Electroencephalograms (EEGs) are a fundamental evaluation in neurology but require special expertise unavailable in many regions of the world. Artificial intelligence (AI) has a potential for addressing these unmet needs. Previous AI models address only limited aspects of EEG interpretation such as distinguishing abnormal from normal or identifying epileptiform activity. A comprehensive, fully automated interpretation of routine EEG based on AI suitable for clinical practice is needed. Objective: To develop and validate an AI model (Standardized Computer-based Organized Reporting of EEG-Artificial Intelligence [SCORE-AI]) with the ability to distinguish abnormal from normal EEG recordings and to classify abnormal EEG recordings into categories relevant for clinical decision-making: epileptiform-focal, epileptiform-generalized, nonepileptiform-focal, and nonepileptiform-diffuse. Design, Setting, and Participants: In this multicenter diagnostic accuracy study, a convolutional neural network model, SCORE-AI, was developed and validated using EEGs recorded between 2014 and 2020. Data were analyzed from January 17, 2022, until November 14, 2022. A total of 30 493 recordings of patients referred for EEG were included into the development data set annotated by 17 experts. Patients aged more than 3 months and not critically ill were eligible. The SCORE-AI was validated using 3 independent test data sets: a multicenter data set of 100 representative EEGs evaluated by 11 experts, a single-center data set of 9785 EEGs evaluated by 14 experts, and for benchmarking with previously published AI models, a data set of 60 EEGs with external reference standard. No patients who met eligibility criteria were excluded. Main Outcomes and Measures: Diagnostic accuracy, sensitivity, and specificity compared with the experts and the external reference standard of patients' habitual clinical episodes obtained during video-EEG recording. Results: The characteristics of the EEG data sets include development data set (N = 30 493; 14 980 men; median age, 25.3 years [95% CI, 1.3-76.2 years]), multicenter test data set (N = 100; 61 men, median age, 25.8 years [95% CI, 4.1-85.5 years]), single-center test data set (N = 9785; 5168 men; median age, 35.4 years [95% CI, 0.6-87.4 years]), and test data set with external reference standard (N = 60; 27 men; median age, 36 years [95% CI, 3-75 years]). The SCORE-AI achieved high accuracy, with an area under the receiver operating characteristic curve between 0.89 and 0.96 for the different categories of EEG abnormalities, and performance similar to human experts. Benchmarking against 3 previously published AI models was limited to comparing detection of epileptiform abnormalities. The accuracy of SCORE-AI (88.3%; 95% CI, 79.2%-94.9%) was significantly higher than the 3 previously published models (P < .001) and similar to human experts. Conclusions and Relevance: In this study, SCORE-AI achieved human expert level performance in fully automated interpretation of routine EEGs. Application of SCORE-AI may improve diagnosis and patient care in underserved areas and improve efficiency and consistency in specialized epilepsy centers.

Neuropsychological effects in children exposed to anticonvulsant monotherapy during gestation: Phenobarbital, carbamazepine, and phenytoin.

OBJECTIVE: Usage during pregnancy of the antiseizure medication (ASM), phenobarbital (PB), carbamazepine (CBZ), and phenytoin (PHT), has been associated with adverse pregnancy outcomes. While morphological effects on offspring are well-documented, inconsistent findings have been reported on neuropsychological development, possibly due to differences in attention to maternal demographics, and other design characteristics. Herein, we report the results of a carefully designed protocol used to examine the effects of gestational monotherapy with PB, CBZ, or PHT upon children's general mental abilities, when compared to age- and gender- matched children born to unexposed women of similar age, education, and socioeconomic status. METHODS: For each ASM, we selected qualifying cases from children born to PB, CBZ, or PHT monotherapy-exposed and unexposed women. Following the application of inclusion, exclusion, and matching criteria, our sample included 34 PB-exposed, 40 PHT-exposed, and 41 CBZ-exposed children along with matched unexposed children for each drug group. Criteria were applied through examination of maternal medical and educational histories, parental socioeconomic characteristics, and child's age and gender. Each child's physical and neuropsychological characteristics were examined, using standardized protocols. We report on the cognitive performance of the children as assessed by the Wechsler Intelligence Scale for Children - III (WISC-III), the leading measure of mental ability in the U.S. RESULTS: An overall mixed model ANOVA of the adjusted performance of the children across all groups controlling for maternal IQ revealed significant effects on verbal IQ, but not full-scale IQ or performance IQ. In the individual drug and unexposed group comparisons, only reduced verbal and full-scale IQ scores in PB-exposed versus matched unexposed children were found. Comparisons between drug groups revealed a significant reduction in verbal IQ and full-scale IQ in PB-exposed versus PHT-exposed children, but not in other drug-drug comparisons. SIGNIFICANCE: These results demonstrate effects on children's mental ability due to prenatal PB exposure, such that analyses adjusted for maternal IQ scores, revealed reduced verbal mental abilities and reduced full-scale IQ scores when scores in exposed children were compared to scores from children of the same age and sex born to demographically similar, healthy unexposed women. When comparisons were made between drug groups, children exposed prenatally to PB performed significantly worse than prenatally PHT-exposed children, but CBZ-exposed children's scores were not significantly different from those of PB or PHT-exposed groups. In light of shared effects on structural teratogenicity, these findings suggest that use of PB monotherapy for the management of seizures during pregnancy may be associated with increased risk in comparison to PHT when neurobehavioral functioning is considered, and that only PB-exposed children have reduced performance compared to matched controls. Attention to these effects is critical in the developing world where use of these older medications remains predominant, and prudent choices can be made to reduce impact on cognitive development.

Microscale Physiological Events on the Human Cortical Surface.

Despite ongoing advances in our understanding of local single-cellular and network-level activity of neuronal populations in the human brain, extraordinarily little is known about their "intermediate" microscale local circuit dynamics. Here, we utilized ultra-high-density microelectrode arrays and a rare opportunity to perform intracranial recordings across multiple cortical areas in human participants to discover three distinct classes of cortical activity that are not locked to ongoing natural brain rhythmic activity. The first included fast waveforms similar to extracellular single-unit activity. The other two types were discrete events with slower waveform dynamics and were found preferentially in upper cortical layers. These second and third types were also observed in rodents, nonhuman primates, and semi-chronic recordings from humans via laminar and Utah array microelectrodes. The rates of all three events were selectively modulated by auditory and electrical stimuli, pharmacological manipulation, and cold saline application and had small causal co-occurrences. These results suggest that the proper combination of high-resolution microelectrodes and analytic techniques can capture neuronal dynamics that lay between somatic action potentials and aggregate population activity. Understanding intermediate microscale dynamics in relation to single-cell and network dynamics may reveal important details about activity in the full cortical circuit.

Ambulatory EEG Monitoring, Reviewing, and Interpreting.

SUMMARY: This section of this volume of the Journal of Neurophysiology's review of ambulatory EEG (aEEG) is an overview of the uses of the technique, the clinicians who most often use it, and why they use it. It looks at the use of aEEG from the perspectives of the "generalist" and the "specialist" and discusses where they have similar interests and where their interest may diverge. It compares the use of aEEG to the more classic inpatient-based epilepsy monitoring unit service and compares the advantages and disadvantages of the two approaches (aEEG vs. epilepsy monitoring unit). This section then reviews how aEEGs are clinically interpreted based on a best practices approach and how aEEGs can be modified to address recording problems that, until recently, have been approached only in an epilepsy monitoring unit environment. Finally, this section addresses the qualifications necessary for the proper interpretation and reporting of aEEG studies. At the end of this section, there are five case examples using aEEG techniques that demonstrate various aspects covered in this section. These case reports demonstrate not only the utility of aEEG but also show how aEEG recordings can be integrated into the care of complex clinical situations that are frequently encountered not only by the practicing generalists but also by epilepsy specialists who practices out of a comprehensive epilepsy program.

Electroencephalography: basic biophysical and technological aspects important for clinical applications.

Electroencephalography (EEG) is the most commonly used functional investigative method in epilepsy. The goal of this educational review paper is to summarize the most important aspects related to the biophysical phenomena of EEG signal generation and the technical features that a clinician needs to understand in order to read and interpret EEGs. We explain the EEG electrodes and recording arrays, amplifiers, filters, analogue-to-digital conversion and signal display. We describe the advantages and disadvantages of the different montages and the indications for the various types of EEG recordings and provocative maneuvers. We explain how to use topographic maps to estimate the source of the cortical generator.

Standardized computer-based organized reporting of EEG: SCORE - Second version.

Standardized terminology for computer-based assessment and reporting of EEG has been previously developed in Europe. The International Federation of Clinical Neurophysiology established a taskforce in 2013 to develop this further, and to reach international consensus. This work resulted in the second, revised version of SCORE (Standardized Computer-based Organized Reporting of EEG), which is presented in this paper. The revised terminology was implemented in a software package (SCORE EEG), which was tested in clinical practice on 12,160 EEG recordings. Standardized terms implemented in SCORE are used to report the features of clinical relevance, extracted while assessing the EEGs. Selection of the terms is context sensitive: initial choices determine the subsequently presented sets of additional choices. This process automatically generates a report and feeds these features into a database. In the end, the diagnostic significance is scored, using a standardized list of terms. SCORE has specific modules for scoring seizures (including seizure semiology and ictal EEG patterns), neonatal recordings (including features specific for this age group), and for Critical Care EEG Terminology. SCORE is a useful clinical tool, with potential impact on clinical care, quality assurance, data-sharing, research and education.

Laminar profile of spontaneous and evoked theta: Rhythmic modulation of cortical processing during word integration.

Theta may play a central role during language understanding and other extended cognitive processing, providing an envelope for widespread integration of participating cortical areas. We used linear microelectrode arrays in epileptics to define the circuits generating theta in inferotemporal, perirhinal, entorhinal, prefrontal and anterior cingulate cortices. In all locations, theta was generated by excitatory current sinks in middle layers which receive predominantly feedforward inputs, alternating with sinks in superficial layers which receive mainly feedback/associative inputs. Baseline and event-related theta were generated by indistinguishable laminar profiles of transmembrane currents and unit-firing. Word presentation could reset theta phase, permitting theta to contribute to late event-related potentials, even when theta power decreases relative to baseline. Limited recordings during sentence reading are consistent with rhythmic theta activity entrained by a given word modulating the neural background for the following word. These findings show that theta occurs spontaneously, and can be momentarily suppressed, reset and synchronized by words. Theta represents an alternation between feedforward/divergent and associative/convergent processing modes that may temporally organize sustained processing and optimize the timing of memory formation. We suggest that words are initially encoded via a ventral feedforward stream which is lexicosemantic in the anteroventral temporal lobe; its arrival may trigger a widespread theta rhythm which integrates the word within a larger context.

Information theoretic measures of network coordination in high-frequency scalp EEG reveal dynamic patterns associated with seizure termination.

How a seizure terminates is still under-studied and, despite its clinical importance, remains an obscure phase of seizure evolution. Recent studies of seizure-related scalp EEGs at frequencies >100 Hz suggest that neural activity, in the form of oscillations and/or neuronal network interactions, may play an important role in preictal/ictal seizure evolution (Andrade-Valenca et al., 2011; Stamoulis et al., 2012). However, the role of high-frequency activity in seizure termination, is unknown, if it exists at all. Using information theoretic measures of network coordination, this study investigated ictal and immediate postictal neurodynamic interactions encoded in scalp EEGs from a relatively small sample of 8 patients with focal epilepsy and multiple seizures originating in temporal and/or frontal brain regions, at frequencies ≤ 100 Hz and >100 Hz, respectively. Despite some heterogeneity in the dynamics of these interactions, consistent patterns were also estimated. Specifically, in several seizures, linear or non-linear increase in high-frequency neuronal coordination during ictal intervals, coincided with a corresponding decrease in coordination at frequencies <100 Hz, suggesting a potential interference role of high-frequency activity, to disrupt abnormal ictal synchrony at lower frequencies. These changes in network synchrony started at least 20-30s prior to seizure offset, depending on the seizure duration. Opposite patterns were estimated at frequencies ≤ 100 Hz in several seizures. These results raise the possibility that high-frequency interference may occur in the form of progressive network coordination during the ictal interval, which continues during the postictal interval. This may be one of several possible mechanisms that facilitate seizure termination. In fact, inhibition of pairwise interactions between EEGs by other signals in their spatial neighborhood, quantified by negative interaction information, was estimated at frequencies ≤ 100 Hz, at least in some seizures.

High-frequency neuronal network modulations encoded in scalp EEG precede the onset of focal seizures.

Modulations of neuronal network interactions by seizure precursors are only partially understood and difficult to measure, in part due to inherent intra- and inter-patient seizure heterogeneities and EEG variability. This study investigated preictal neuromodulations associated with seizures originating in the temporal and/or frontal lobes, using information theoretic parameters estimated from awake scalp EEGs in two frequency ranges, ≤100 Hz and >100 Hz, respectively. Seizure-related activity at high frequencies has not been extensively estimated in awake scalp EEGs. Based on the statistical similarity of preictal and ictal information parameters, preictal network interactions appeared to be specifically modulated at frequencies >100 Hz, but not at lower frequencies. The dynamics of these parameters varied distinctly according to the origin of seizure onset (temporal versus frontal). Although preliminary, and based on a small patient sample for which the potential heterogeneity of multiple anticonvulsive medications was difficult to control, these results suggest that preictal modulations may be estimated from high-frequency scalp EEGs using directional information measures with high specificity to ictal events, and may thus be promising for improving seizure prediction.

Survivors and nonsurvivors of very prolonged status epilepticus.

Several studies have shown reliable predictive factors for outcome in status epilepticus (SE), especially etiology and presentation in coma. Duration of SE is predictive, but probably only in the first few hours, and there have been many reports of patients treated successfully for SE lasting many days or weeks. Nevertheless, there are many other patients with SE treated for prolonged periods without success, sometimes apparently futilely. We compared clinical features of 10 survivors of prolonged SE with those of a matched cohort treated for similarly prolonged episodes but unsuccessfully, looking for exceptions to known predictive factors. Multiple medical problems (i.e., etiologies) and coma on presentation were confirmed as predictors of a poor outcome. Analysis of individual exceptions to these predictors showed that age, overall background health, and family input on the value of prolonged treatment, on the one hand, and earlier epilepsy plus rapid and accurate diagnosis and treatment, on the other, contributed to results different from what would have been expected.

Artifact: recording EEG in special care units.

Artifacts may be obtained during routine recording but are more common in special care units (SCUs) outside of the EEG laboratory, where complex electrical currents are present that create a "hostile" environment. Special care units include the epilepsy monitoring unit, neurologic intensive care unit, and operating room, where artifact is present in virtually every recording, increasing with prolonged use. Nonepileptic attacks treated as epileptic seizures have been incorrectly diagnosed and treated due to a misinterpreted EEG. The recent emergence of continuous EEG as a neurophysiologic surrogate for brain function in the neurologic intensive care unit and operating room has also brought a greater amount and new types of EEG artifact. The artifacts encountered in special care units during continuous EEG are becoming more complex and may have adverse therapeutic implications. Our knowledge of artifact needs to parallel our growth in technology to avoid the pitfalls that may be incurred during visual analysis of the EEG.

Artifact and recording concepts in EEG.

Artifact is present when electrical potentials that are not brain derived are recorded on the EEG and is commonly encountered during interpretation. Many artifacts obscure the tracing, while others reflect physiologic functions that are crucial for routine visual analysis. Both physiologic and nonphysiologic sources of artifact may act as source of confusion with abnormality and lead to misinterpretation. Identifying the mismatch between potentials that are generated by the brain from activity that does not conform to a realistic head model is the foundation for recognizing artifact. Electroencephalographers are challenged with the task of correct interpretations among the many artifacts that could potentially be misleading, resulting in an incorrect diagnosis and treatment that may adversely impact patient care. Despite advances in digital EEG, artifact identification, recognition, and elimination are essential for correct interpretation of the EEG. The authors discuss recording concepts for interpreting EEG that contains artifact.

Heterogeneous neuronal firing patterns during interictal epileptiform discharges in the human cortex.

Epileptic cortex is characterized by paroxysmal electrical discharges. Analysis of these interictal discharges typically manifests as spike-wave complexes on electroencephalography, and plays a critical role in diagnosing and treating epilepsy. Despite their fundamental importance, little is known about the neurophysiological mechanisms generating these events in human focal epilepsy. Using three different systems of microelectrodes, we recorded local field potentials and single-unit action potentials during interictal discharges in patients with medically intractable focal epilepsy undergoing diagnostic workup for localization of seizure foci. We studied 336 single units in 20 patients. Ten different cortical areas and the hippocampus, including regions both inside and outside the seizure focus, were sampled. In three of these patients, high density microelectrode arrays simultaneously recorded between 43 and 166 single units from a small (4 mm x 4 mm) patch of cortex. We examined how the firing rates of individual neurons changed during interictal discharges by determining whether the firing rate during the event was the same, above or below a median baseline firing rate estimated from interictal discharge-free periods (Kruskal-Wallis one-way analysis, P<0.05). Only 48% of the recorded units showed such a modulation in firing rate within 500 ms of the discharge. Units modulated during the discharge exhibited significantly higher baseline firing and bursting rates than unmodulated units. As expected, many units (27% of the modulated population) showed an increase in firing rate during the fast segment of the discharge (+ or - 35 ms from the peak of the discharge), while 50% showed a decrease during the slow wave. Notably, in direct contrast to predictions based on models of a pure paroxysmal depolarizing shift, 7.7% of modulated units recorded in or near the seizure focus showed a decrease in activity well ahead (0-300 ms) of the discharge onset, while 12.2% of units increased in activity in this period. No such pre-discharge changes were seen in regions well outside the seizure focus. In many recordings there was also a decrease in broadband field potential activity during this same pre-discharge period. The different patterns of interictal discharge-modulated firing were classified into more than 15 different categories. This heterogeneity in single unit activity was present within small cortical regions as well as inside and outside the seizure onset zone, suggesting that interictal epileptiform activity in patients with epilepsy is not a simple paroxysm of hypersynchronous excitatory activity, but rather represents an interplay of multiple distinct neuronal types within complex neuronal networks.

The relationship of interictal epileptiform discharges to clinical epilepsy severity: a study of routine electroencephalograms and review of the literature.

Electroencephalograms are widely used to detect interictal epileptiform discharges (IEDs) in patients with a known history of seizures. However, previous studies have not found a consistent association between the presence or frequency of IEDs and clinical epilepsy severity, possibly because of differences in subject characteristics and recording techniques. We sought to investigate this relationship in a population and setting reflective of the most common clinical usage. We analyzed electroencephalograms and clinical records of all consenting patients with a history of at least two presumed focal-onset seizures who presented for routine electroencephalograms recording over 1-year time in an academic neurophysiology laboratory (n = 129). Despite adequate statistical power, we did not find an association between the presence or absence of IEDs or IED frequency and the most recently determined seizure frequency (median, 4 per year). A higher IED incidence was seen in subjects with longer epilepsy duration (P = 0.04). Neither IED incidence nor frequency (median, 10.0 per hour) correlated with age or antiepileptic drug use. Our results differ from those of some previous studies, most of which focused on more narrow subject populations, suggesting that the patient's clinical circumstances must be taken into account before assuming the utility of IEDs on routine electroencephalography in predicting epilepsy severity.

A comparison of anastrozole and testosterone versus placebo and testosterone for treatment of sexual dysfunction in men with epilepsy and hypogonadism.

Hyposexuality is commonly associated with low bioavailable testosterone (BAT) and relative estradiol elevation in men with epilepsy. This prospective, randomized, double-blind trial compared the effects of depotestosterone+the aromatase inhibitor anastrozole (T-A) versus depotestosterone+placebo (T-P) on sexual function, hormone levels, mood, and seizure frequency in men with epilepsy. Forty men with focal epilepsy, hyposexuality, and hypogonadism were randomized 1:1 to two groups (T-A or T-P) for a 3-month treatment trial of depotestosterone+either anastrozole or matching placebo. Outcomes included both efficacy and safety measures. Normalization of sexual function (S-score) occurred with greater frequency in the T-A (72.2%) than in the T-P (47.4%) group, but the difference was not statistically significant. T-A resulted in significantly lower estradiol levels and S-scores correlated inversely with estradiol levels at baseline and during treatment. Beck Depression Inventory II (BDI-II) scores improved significantly in both groups and changes in S-score correlated inversely with changes in BDI-II score. Changes in seizure frequency correlated with changes in BDI-II score. Seizure frequency decreased with both treatments and showed significant correlations with estradiol levels. Triglyceride levels increased with T-P and decreased with T-A. The difference in triglyceride changes between the two treatments was significant and correlated with changes in estradiol levels. Significant correlations between estradiol levels and S-scores, as well as seizure outcomes and triglyceride levels, suggest further study regarding a potential role for anastrozole in the treatment of men with epilepsy who have hyposexuality and hypogonadism.

Experimental validation of the influence of white matter anisotropy on the intracranial EEG forward solution.

Forward solutions with different levels of complexity are employed for localization of current generators, which are responsible for the electric and magnetic fields measured from the human brain. The influence of brain anisotropy on the forward solution is poorly understood. The goal of this study is to validate an anisotropic model for the intracranial electric forward solution by comparing with the directly measured 'gold standard'. Dipolar sources are created at known locations in the brain and intracranial electroencephalogram (EEG) is recorded simultaneously. Isotropic models with increasing level of complexity are generated along with anisotropic models based on Diffusion tensor imaging (DTI). A Finite Element Method based forward solution is calculated and validated using the measured data. Major findings are (1) An anisotropic model with a linear scaling between the eigenvalues of the electrical conductivity tensor and water self-diffusion tensor in brain tissue is validated. The greatest improvement was obtained when the stimulation site is close to a region of high anisotropy. The model with a global anisotropic ratio of 10:1 between the eigenvalues (parallel: tangential to the fiber direction) has the worst performance of all the anisotropic models. (2) Inclusion of cerebrospinal fluid as well as brain anisotropy in the forward model is necessary for an accurate description of the electric field inside the skull. The results indicate that an anisotropic model based on the DTI can be constructed non-invasively and shows an improved performance when compared to the isotropic models for the calculation of the intracranial EEG forward solution.

Advances in the application of technology to epilepsy: the CIMIT/NIO Epilepsy Innovation Summit.

In 2008, a group of clinicians, scientists, engineers, and industry representatives met to discuss advances in the application of engineering technologies to the diagnosis and treatment of patients with epilepsy. The presentations also provided a guide for further technological development, specifically in the evaluation of patients for epilepsy surgery, seizure onset detection and seizure prediction, intracranial treatment systems, and extracranial treatment systems. This article summarizes the discussions and demonstrates that cross-disciplinary interactions can catalyze collaborations between physicians and engineers to address and solve many of the pressing unmet needs in epilepsy.