Review of epilepsy care in the Democratic Republic of the Congo.

Epilepsy imposes a substantial burden on the Democratic Republic of Congo (DRC). These challenges encompass the lack of comprehensive disease surveillance, an unresolved understanding of its pathophysiology, economic barriers limiting access to essential care, the absence of epilepsy surgical capabilities, and deeply ingrained societal stigmas. Notably, the national prevalence of epilepsy remains undetermined, with research primarily concentrating on infectious factors like Onchocerca volvulus, leaving other potential causes underexplored. Most patients lack insurance, incurring out-of-pocket expenses that often lead them to opt for traditional medicine rather than clinical care. Social stigma, perpetuated by common misconceptions, intensifies the social isolation experienced by individuals living with epilepsy. Additionally, surgical interventions are unavailable, and the accessibility of anti-seizure medications and healthcare infrastructure remains inadequate. Effectively tackling these interrelated challenges requires a multifaceted approach, including conducting research into region-specific factors contributing to epilepsy, increasing healthcare funding, subsidizing the costs of treatment, deploying mobile tools for extensive screening, launching awareness campaigns to dispel myths and reduce stigma, and promoting collaborations between traditional healers and medical practitioners to enhance local understanding and epilepsy management. Despite the difficulties, significant progress can be achieved through sustained and compassionate efforts to understand and eliminate the barriers faced by epilepsy patients in the region. This review outlines essential steps for alleviating the epilepsy burden in the DRC. PLAIN LANGUAGE SUMMARY: There are not enough resources to treat epilepsy in the DRC. PWEs struggle with stigma and the lack of money. Many of them still use traditional medicine for treatment and hold wrong beliefs about epilepsy. That is why there is a need for more resources to make the lives of PWEs better in the DRC.

Functional and anatomical connectivity predict brain stimulation's mnemonic effects.

Closed-loop direct brain stimulation is a promising tool for modulating neural activity and behavior. However, it remains unclear how to optimally target stimulation to modulate brain activity in particular brain networks that underlie particular cognitive functions. Here, we test the hypothesis that stimulation's behavioral and physiological effects depend on the stimulation target's anatomical and functional network properties. We delivered closed-loop stimulation as 47 neurosurgical patients studied and recalled word lists. Multivariate classifiers, trained to predict momentary lapses in memory function, triggered the stimulation of the lateral temporal cortex (LTC) during the study phase of the task. We found that LTC stimulation specifically improved memory when delivered to targets near white matter pathways. Memory improvement was largest for targets near white matter that also showed high functional connectivity to the brain's memory network. These targets also reduced low-frequency activity in this network, an established marker of successful memory encoding. These data reveal how anatomical and functional networks mediate stimulation's behavioral and physiological effects, provide further evidence that closed-loop LTC stimulation can improve episodic memory, and suggest a method for optimizing neuromodulation through improved stimulation targeting.

Functional and anatomical connectivity predict brain stimulation's mnemonic effects.

Closed-loop direct brain stimulation is a promising tool for modulating neural activity and behavior. However, it remains unclear how to optimally target stimulation to modulate brain activity in particular brain networks that underlie particular cognitive functions. Here, we test the hypothesis that stimulation's behavioral and physiological effects depend on the stimulation target's anatomical and functional network properties. We delivered closed-loop stimulation as 47 neurosurgical patients studied and recalled word lists. Multivariate classifiers, trained to predict momentary lapses in memory function, triggered stimulation of the lateral temporal cortex (LTC) during the study phase of the task. We found that LTC stimulation specifically improved memory when delivered to targets near white matter pathways. Memory improvement was largest for targets near white matter that also showed high functional connectivity to the brain's memory network. These targets also reduced low-frequency activity in this network, an established marker of successful memory encoding. These data reveal how anatomical and functional networks mediate stimulation's behavioral and physiological effects, provide further evidence that closed-loop LTC stimulation can improve episodic memory, and suggest a method for optimizing neuromodulation through improved stimulation targeting.

Disparities in adherence and emergency department utilization among people with epilepsy: A machine learning approach.

PURPOSE: We used a machine learning approach to identify the combinations of factors that contribute to lower adherence and high emergency department (ED) utilization. METHODS: Using Medicaid claims, we identified adherence to anti-seizure medications and the number of ED visits for people with epilepsy in a 2-year follow up period. We used three years of baseline data to identify demographics, disease severity and management, comorbidities, and county-level social factors. Using Classification and Regression Tree (CART) and random forest analyses we identified combinations of baseline factors that predicted lower adherence and ED visits. We further stratified these models by race and ethnicity. RESULTS: From 52,175 people with epilepsy, the CART model identified developmental disabilities, age, race and ethnicity, and utilization as top predictors of adherence. When stratified by race and ethnicity, there was variation in the combinations of comorbidities including developmental disabilities, hypertension, and psychiatric comorbidities. Our CART model for ED utilization included a primary split among those with previous injuries, followed by anxiety and mood disorders, headache, back problems, and urinary tract infections. When stratified by race and ethnicity we saw that for Black individuals headache was a top predictor of future ED utilization although this did not appear in other racial and ethnic groups. CONCLUSIONS: ASM adherence differed by race and ethnicity, with different combinations of comorbidities predicting lower adherence across racial and ethnic groups. While there were not differences in ED use across races and ethnicity, we observed different combinations of comorbidities that predicted high ED utilization.

Biomarker-guided neuromodulation aids memory in traumatic brain injury.

Traumatic brain injury (TBI) is a leading cause of cognitive disability in adults, often characterized by marked deficits in episodic memory and executive function. Prior studies have found that direct electrical stimulation of the temporal cortex yielded improved memory in epilepsy patients, but it is not clear if these results generalize to patients with a specific history of TBI. Here we asked whether applying closed-loop, direct electrical stimulation to lateral temporal cortex could reliably improve memory in a TBI cohort. Among a larger group of patients undergoing neurosurgical evaluation for refractory epilepsy, we recruited a subset of patients with a history of moderate-to-severe TBI. By analyzing neural data from indwelling electrodes as patients studied and recalled lists of words, we trained personalized machine-learning classifiers to predict momentary fluctuations in mnemonic function in each patient. We subsequently used these classifiers to trigger high-frequency stimulation of the lateral temporal cortex (LTC) at moments when memory was predicted to fail. This strategy yielded a 19% boost in recall performance on stimulated as compared with non-stimulated lists (P = 0.012). These results provide a proof-of-concept for using closed-loop stimulation of the brain in treatment of TBI-related memory impairment.

Racial and Ethnic Differences in Antiseizure Medications Among People With Epilepsy on Medicaid: A Case of Potential Inequities.

Background and Objectives: Being on a newer, second-, and third-generation antiseizure medication (ASM) may represent an important marker of quality of care for people with epilepsy. We sought to examine whether there were racial/ethnic differences in their use. Methods: Using Medicaid claims data, we identified the type and number of ASMs, as well as the adherence, for people with epilepsy over a 5-year period (2010-2014). We used multilevel logistic regression models to examine the association between newer-generation ASMs and adherence. We then examined whether there were racial/ethnic differences in ASM use in models adjusted for demographics, utilization, year, and comorbidities. Results: Among 78,534 adults with epilepsy, 17,729 were Black, and 9,376 were Hispanic. Overall, 25.6% were on older ASMs, and being solely on second-generation ASMs during the study period was associated with better adherence (adjusted odds ratio: 1.17, 95% confidence interval [CI]: 1.11-1.23). Those who saw a neurologist (3.26, 95% CI: 3.13-3.41) or who were newly diagnosed (1.29, 95% CI: 1.16-1.42) had higher odds of being on newer ASMs. Importantly, Black (0.71, 95% CI: 0.68-0.75), Hispanic (0.93, 95% CI: 0.88-0.99), and Native Hawaiian and Other Pacific Island individuals (0.77, 95% CI: 0.67-0.88) had lower odds of being on newer ASMs when compared with White individuals. Discussion: Generally, racial and ethnic minoritized people with epilepsy have lower odds of being on newer-generation ASMs. Greater adherence by people who were only on newer ASMs, their greater use among people seeing a neurologist, and the opportunity of a new diagnosis point to actionable leverage points for reducing inequities in epilepsy care.

Accelerated long-term forgetting of recall and recognition memory in people with epilepsy.

PURPOSE: Persons with epilepsy (PWE) report memory deficits as one of the most distressing aspects of their disorder. Recently, a long-term memory deficit known as Accelerated Long-Term Forgetting (ALF) has been described in PWE. ALF is characterized by the initial retention of learned information, followed by an accelerated rate of memory decay. However, the rate of ALF varies widely across literature and it is unclear how it impacts different memory retrieval types. The current study aimed to capture the time course of ALF on both free recall and recognition memory using a movie-based task in PWE. METHODS: A sample of 30 PWE and 30 healthy comparison (HC) subjects watched a nature documentary and were tested on their recall and recognition of the film's content immediately after viewing and at delays of 24 hours, 48 hours, and 72 hours. Participants also rated the confidence they had in their recognition memory trial responses. RESULTS: For recall, PWE exhibit ALF at 72 hours (ß = -19.840, SE = 3.743, z(226) = -5.301, p < 0.001). For recognition, PWE had decreased performance compared to controls at the 24-hour (ß = -10.165, SE = 4.174, z(224) = -3.166, p = 0.004), 48-hour (ß = -8.113, SE = 3.701, z(224) = -2.195, p = 0.044), and 72-hour (ß = -10.794, SE = 3.017, z(224) = -3.295, p = 0.003) delays. The PWE group showed positive correlations (tau = 0.165, p < 0.001) between confidence ratings and accuracy, with higher confidence reflecting successful recognition. PWE were 49% less likely to answer either retrieval type correctly at 72 hours (OR 0.51, 95% CI [0.35, 0.74], p < 0.001). Left hemispheric seizure onset decreased the odds of successful retrieval by 88% (OR 0.12, 95% CI [0.01, 0.42], p = 0.019). CONCLUSIONS: These findings provide evidence of ALF in PWE, with a differential impact on recall and recognition memory. This further supports the call to include ALF assessments in standard memory evaluations in PWE. Additionally, identifying the neural correlates of ALF in the future will be important in developing targeted therapies to alleviate the burden of memory impairment for PWE.

Functional Reorganization of the Mesial Frontal Premotor Cortex in Patients With Supplementary Motor Area Seizures.

BACKGROUND: Direct cortical stimulation of the mesial frontal premotor cortex, including the supplementary motor area (SMA), is challenging in humans. Limited access to these brain regions impedes understanding of human premotor cortex functional organization and somatotopy. OBJECTIVE: To test whether seizure onset within the SMA was associated with functional remapping of mesial frontal premotor areas in a cohort of patients with epilepsy who underwent awake brain mapping after implantation of interhemispheric subdural electrodes. METHODS: Stimulation trials from 646 interhemispheric subdural electrodes were analyzed and compared between patients who had seizure onset in the SMA (n = 13) vs patients who had seizure onset outside of the SMA (n = 12). 1:1 matching with replacement between SMA and non-SMA data sets was used to ensure similar spatial distribution of electrodes. Centroids and 95% confidence regions were computed for clustered head, trunk, upper extremity, lower extremity, and vision responses. A generalized linear mixed-effects model was used to test for significant differences in the resulting functional maps. Clinical, radiographic, and histopathologic data were reviewed. RESULTS: After analyzing direct cortical stimulation trials from interhemispheric electrodes, we found significant displacement of the head and trunk responses in SMA compared with non-SMA patients ( P < .01 for both). These differences remained significant after accounting for structural lesions, preexisting motor deficits, and seizure outcome. CONCLUSION: The somatotopy of the mesial frontal premotor regions is significantly altered in patients who have SMA-onset seizures compared with patients who have seizure onset outside of the SMA, suggesting that functional remapping can occur in these brain regions.

A framework for health equity in people living with epilepsy.

Epilepsy is a disease where disparities and inequities in risk and outcomes are complex and multifactorial. While most epilepsy research to date has identified several key areas of disparities, we set out to provide a multilevel life course model of epilepsy development, diagnosis, treatment, and outcomes to highlight how these disparities represent true inequities. Our piece also presents three hypothetical cases that highlight how the solutions to address inequities may vary across the lifespan. We then identify four key domains (structural, socio-cultural, health care, and physiological) that contribute to the persistence of inequities in epilepsy risk and outcomes in the United States. Each of these domains, and their core components in the context of epilepsy, are reviewed and discussed. Further, we highlight the connection between domains and key areas of intervention to strive towards health equity. The goal of this work is to highlight these domains while also providing epilepsy researchers and clinicians with broader context of how their work fits into health equity.

Epilepsy and COVID 2021.

Coronavirus 19 (COVID-19) has infected over 400 million people worldwide. Although COVID-19 causes predominantly respiratory symptoms, it can affect other organs including the brain, producing neurological symptoms. People with epilepsy (PWE) have been particularly impacted during the pandemic with decreased access to care, increased stress, and worsening seizures in up to 22% of them probably due to multiple factors. COVID-19 vaccines were produced in a record short time and have yielded outstanding protection with very rare serious side effects. Studies have found that COVID-19 vaccination does not increase seizures in the majority of PWE. COVID-19 does not produce a pathognomonic EEG or seizure phenotype, but rather 1 that can be seen in other types of encephalopathy. COVID-19 infection and its complications can lead to seizures, status epilepticus and post-COVID inflammatory syndrome with potential multi-organ damage in people without pre-existing epilepsy. The lack of access to care during the pandemic has forced patients and doctors to rapidly implement telemedicine. The use of phone videos and smart telemedicine are helping to treat patients during this pandemic and are becoming standard of care. Investment in infrastructure is important to make sure patients can have access to care even during a pandemic.

Premature discontinuation among individuals with epilepsy participating in epilepsy self-management research interventions.

OBJECTIVE: The Managing Epilepsy Well (MEW) Network conducts epilepsy self-management (ESM) intervention development, testing, and archival clinical trials data analyses in the MEW Network Integrated Database (MEW-DB). However, not all trial participants fully benefit from ESM due to prematurely discontinuing program participation. This analysis sought to identify demographic and clinical predictors of premature discontinuation (PD) of ESM interventions available in the MEW-DB. METHODS: Data from prior studies were included if: 1) they were prospective trials testing an ESM intervention, 2) included baseline assessment and at least one post-intervention assessment, and 3) included data on PD. Dependent variables were all-cause PD, categorized as a binary variable (yes/no) and time to PD for the intervals between baseline and follow-up visit 1 (V1; approximately week 12) and visit two (V2; approximately week 24). Multivariable Cox proportional hazard models were used to identify factors affecting PD time-point. Explanatory variables included age, gender, race/ethnicity, education, employment, income, marital status, psychiatric comorbidities, depressive severity, anxiety symptoms, self-efficacy, number of anti-seizure medications (ASMs), health status, seizure frequency, and study design. RESULTS: Six prior MEW-DB studies were included, consisting of 627 people, where 624 were assigned to ESM or to control. PD among randomized individuals was 14.3 % by V1 and 15.7 % by V2. Predictors for V1 PD were treatment (ESM) vs. control arm, more severe depressive symptoms and having schizophrenia. Predictors for V2 PD were younger age, white race, more severe depressive symptoms and having schizophrenia. SIGNIFICANCE: While ESM approaches can improve multiple health outcomes among people with epilepsy, nearly one in six individuals prematurely discontinues their program. These findings suggest that ESM interventionists need to be particularly attentive to program retention over the first 3 months after ESM initiation. Younger people with epilepsy, those who self-identify as white, those with schizophrenia, and/or more severe depressive symptoms may need additional support for engagement.

Practical considerations in epilepsy neurostimulation.

Neuromodulation is a key therapeutic tool for clinicians managing patients with drug-resistant epilepsy. Multiple devices are available with long-term follow-up and real-world experience. The aim of this review is to give a practical summary of available neuromodulation techniques to guide the selection of modalities, focusing on patient selection for devices, common approaches and techniques for initiation of programming, and outpatient management issues. Vagus nerve stimulation (VNS), deep brain stimulation of the anterior nucleus of the thalamus (DBS-ANT), and responsive neurostimulation (RNS) are all supported by randomized controlled trials that show safety and a significant impact on seizure reduction, as well as a suggestion of reduction in the risk of sudden unexplained death in epilepsy (SUDEP). Significant seizure reductions are observed after 3 months for DBS, RNS, and VNS in randomized controlled trials, and efficacy appears to improve with time out to 7 to 10 years of follow-up for all modalities, albeit in uncontrolled follow-up or retrospective studies. A significant number of patients experience seizure-free intervals of 6 months or more with all three modalities. Number and location of epileptogenic foci are important factors affecting efficacy, and together with comorbidities such as severe mood or sleep disorders, may influence the choice of modality. Programming has evolved-DBS is typically initiated at lower current/voltage than used in the pivotal trial, whereas target charge density is lower with RNS, however generalizable optimal parameters are yet to be defined. Noninvasive brain stimulation is an emerging stimulation modality, although it is currently not used widely. In summary, clinical practice has evolved from those established in pivotal trials. Guidance is now available for clinicians who wish to expand their approach, and choice of neuromodulation technique may be tailored to individual patients based on their epilepsy characteristics, risk tolerance, and preferences.

Temporal lobe interictal spikes disrupt encoding and retrieval of verbal memory: A subregion analysis.

OBJECTIVE: The medial temporal lobe (MTL) encodes and recalls memories and can be a predominant site for interictal spikes (IS) in patients with focal epilepsy. It is unclear whether memory deficits are due to IS in the MTL producing a transient decline. Here, we investigated whether IS in the MTL subregions and lateral temporal cortex impact episodic memory encoding and recall. METHODS: Seventy-eight participants undergoing presurgical evaluation for medically refractory focal epilepsy with depth electrodes placed in the temporal lobe participated in a verbal free recall task. IS were manually annotated during the pre-encoding, encoding, and recall epochs. We examined the effect of IS on word recall using mixed-effects logistic regression. RESULTS: IS in the left hippocampus (odds ratio [OR] = .73, 95% confidence interval [CI] = .63-.84, p < .001) and left middle temporal gyrus (OR = .46, 95% CI = .27-.78, p < .05) during word encoding decreased subsequent recall performance. Within the left hippocampus, this effect was specific for area CA1 (OR = .76, 95% CI = .66-.88, p < .01) and dentate gyrus (OR = .74, 95% CI = .62-.89, p < .05). IS in other MTL subregions or inferior and superior temporal gyrus and IS occurring during the prestimulus window did not affect word encoding (p > .05). IS during retrieval in right hippocampal (OR = .22, 95% CI = .08-.63, p = .01) and parahippocampal regions (OR = .24, 95% CI = .07-.8, p < .05) reduced the probability of recalling a word. SIGNIFICANCE: IS in medial and lateral temporal cortex contribute to transient memory decline during verbal episodic memory.

Evoked responses to note onsets and phrase boundaries in Mozart's K448.

Understanding the neural correlates of perception of hierarchical structure in music presents a direct window into auditory organization. To examine the hypothesis that high-level and low-level structures-i.e. phrases and notes-elicit different neural responses, we collected intracranial electroencephalography (iEEG) data from eight subjects during exposure to Mozart's K448 and directly compared Event-related potentials (ERPs) due to note onsets and those elicited by phrase boundaries. Cluster-level permutation tests revealed that note-onset-related ERPs and phrase-boundary-related ERPs were significantly different at [Formula: see text], 200, and 450 ms relative to note onset and phrase markers. We also observed increased activity in frontal brain regions when processing phrase boundaries. We relate these observations to (1) a process which syntactically binds notes together hierarchically to form larger phrases; (2) positive emotions induced by successful prediction of forthcoming phrase boundaries and violations of melodic expectations at phrase boundaries.

Surgical Treatments for Epilepsy.

PURPOSE OF REVIEW: More than 20 new antiseizure medications have been approved by the US Food and Drug Administration (FDA) in the past 3 decades; however, outcomes in newly diagnosed epilepsy have not improved, and epilepsy remains drug resistant in up to 40% of patients. Evidence supports improved seizure outcomes and quality of life in those who have undergone epilepsy surgery, but epilepsy surgery remains underutilized. This article outlines indications for epilepsy surgery, describes the presurgical workup, and summarizes current available surgical approaches. RECENT FINDINGS: Class I evidence has demonstrated the superiority of resective surgery compared to medical therapy for seizure control and quality of life in patients with drug-resistant epilepsy. The use of minimally invasive options, such as laser interstitial thermal therapy and stereotactic radiosurgery, are alternatives to resective surgery in well-selected patients. Neuromodulation techniques, such as responsive neurostimulation, deep brain stimulation, and vagus nerve stimulation, offer a suitable alternative, especially in those where resective surgery is contraindicated or where patients prefer nonresective surgery. Although neuromodulation approaches reduce seizure frequency, they are less likely to be associated with seizure freedom than resective surgery. SUMMARY: Appropriate patients with drug-resistant epilepsy benefit from epilepsy surgery. If two well-chosen and tolerated medication trials do not achieve seizure control, referral to a comprehensive epilepsy center for a thorough presurgical workup and discussion of surgical options is appropriate. Mounting Class I evidence supports a significantly higher chance of stopping disabling seizures with surgery than with further medication trials.

Supervised machine learning to predict reduced depression severity in people with epilepsy through epilepsy self-management intervention.

OBJECTIVE: To develop a classifier that predicts reductions in depression severity in people with epilepsy after participation in an epilepsy self-management intervention. METHODS: Ninety-three people with epilepsy from three epilepsy self-management randomized controlled trials from the Managing Epilepsy Well (MWE) Network integrated research database met the inclusion criteria. Supervised machine learning algorithms were utilized to develop prediction models for changes in self-reported depression symptom severity. Features considered by the machine learning classifiers include age, gender, race, ethnicity, education, study type, baseline quality of life, and baseline depression symptom severity. The models were trained and evaluated on their ability to predict clinically meaningful improvement (i.e., a reduction of greater than three points on the nine-item Patient Health Questionnaire (PHQ-9)) between baseline and follow-up (<=12 weeks) depression scores. Models tested were a Multilayer Perceptron (ML), Random Forest (RF), Support Vector Machine (SVM), Logistic Regression with Stochastic Gradient Descent (SGD), K-nearest Neighbors (KNN), and Gradient Boosting (GB). A separate, outside dataset of 41 people with epilepsy was used in a validation exercise to examine the top-performing model's generalizability and performance with external data. RESULTS: All six classifiers performed better than our baseline mode classifier. Support Vector Machine had the best overall performance (average area under the curve [AUC] = 0.754, highest subpopulation AUC = 0.963). Our analysis of the SVM features revealed that higher baseline depression symptom severity, study type (i.e., intervention program goals), higher baseline quality of life, and race had the strongest influence on increasing the likelihood that a subject would experience a clinically meaningful improvement in depression scores. From the validation exercise, our top-performing SVM model performed similarly or better than the average SVM model with the outside dataset (average AUC = 0.887). SIGNIFICANCE: We trained an SVM classifier that offers novel insight into subject-specific features that are important for predicting a clinically meaningful improvement in subjective depression scores after enrollment in a self-management program. We provide evidence for machine learning to select subjects that may benefit most from a self-management program and indicate important factors that self-management programs should collect to develop improved digital tools.

A pilot study assessing the feasibility and acceptability of Project UPLIFT adapted for patients with psychogenic nonepileptic seizures.

BACKGROUND: People with psychogenic nonepileptic seizures (PNES) are at elevated risk of multiple psychiatric comorbidities. Current treatment gaps highlight the need for time-limited, distance-delivered interventions that can be cost-effectively delivered to patients with PNES. Project UPLIFT is a self-management program addressing mood problems in epilepsy that has not previously been adapted for PNES. The purpose of this study was to assess the feasibility and acceptability of UPLIFT for patients with PNES. METHODS: Project UPLIFT was minimally adapted by a multidisciplinary research group at Dartmouth-Hitchcock Medical Center. Participants were recruited into a nonrandomized pilot study and were assessed at baseline, after completing the intervention, and at 1-month follow-up. RESULTS: The intervention was adapted and delivered without additional modification. A total of eight participants enrolled in the trial, and six participated in the first session. All six participants who started UPLIFT completed the intervention, with a 98% attendance record across the eight sessions. At follow-up, 100% reported that UPLIFT provided useful tools for daily life. All participants continued using UPLIFT after completing the program, and 83% felt it helped with seizure management. Preliminary results provide proof-of-concept for future efficacy trials. CONCLUSION: This study provides initial support for the feasibility and acceptability of Project UPLIFT, minimally adapted for patients with PNES. As a time-limited program that can be distance-delivered to groups of patients, UPLIFT may be well-suited for the healthcare environment brought on by the COVID-19 pandemic.

AiED: Artificial intelligence for the detection of intracranial interictal epileptiform discharges.

OBJECTIVE: Deep learning provides an appealing solution for the ongoing challenge of automatically classifying intracranial interictal epileptiform discharges (IEDs). We report results from an automated method consisting of a template-matching algorithm and convolutional neural network (CNN) for the detection of intracranial IEDs ("AiED"). METHODS: 1000 intracranial electroencephalogram (EEG) epochs extracted randomly from 307 subjects with refractory epilepsy were annotated independently by two expert neurophysiologists. These annotated epochs were divided into 1062 two-second epochs with IEDs and 1428 two-second epochs without IEDs, which were transformed into spectrograms prior to training the neural network. The highest performing network was validated on an annotated external test set. RESULTS: The final network had an F1-score of 0.95 (95% CI: 0.91-0.98) and an average Area Under the Receiver Operating Characteristic of 0.98 (95% CI: 0.96-1.00). For the external test set, it showed an overall F1-score of 0.71, correctly identifying 100% of all high-amplitude IED complexes, 96.23% of all high-amplitude isolated IEDs, and 66.15% of all IEDs of atypical morphology. CONCLUSIONS: Template-matching combined with a CNN offers a fast, robust method for detecting intracranial IEDs. SIGNIFICANCE: "AiED" is generalizable and achieves comparable performance to human reviewers; it may support clinical and research EEG analyses.