Epilepsia最新文献

{"title":"Thalamic engagement by epileptic spikes as a mechanism for widespread slow oscillation–spindle decoupling","authors":"Anirudh Wodeyar, Mark A. Kramer, Catherine J. Chu","doi":"10.1111/epi.18482","DOIUrl":"10.1111/epi.18482","url":null,"abstract":"<p>We read with great interest the recent article by Schiller and colleagues, “Widespread Decoupling of Spindles and Slow Waves in Temporal Lobe Epilepsy,”<span><sup>1</sup></span> and were struck by how their findings dovetail with our own observations. In our work “Thalamic Epileptic Spikes Disrupt Sleep Spindles in Patients With Epileptic Encephalopathy,”<span><sup>2</sup></span> we found that when a thalamic epileptic spike co-occurs with a cortical slow oscillation, thalamic spindle incidence is suppressed precisely during the phase at which spindles would ordinarily peak (see Figure 2D in Wodeyar et al.<span><sup>2</sup></span>), a clear decoupling of slow oscillations and spindles due to epileptic spikes in the thalamus. This phenomenon was not limited to epileptic encephalopathies, nor to a specific age range (patients were aged 9–55 years), suggesting that once an epileptic spike engages the thalamus, its disruptive influence on slow oscillation–spindle coupling could be global.</p><p>Employing high-density electroencephalography (EEG) and a comparison between healthy controls and patients with epilepsy, Schiller et al. now demonstrate that the slow oscillation–spindle disruption can indeed extend beyond focal epileptogenic sites in temporal lobe epilepsy, broadening our understanding of how local epileptic pathology can be linked to global network consequences. We note that the authors did not detect a direct relationship between epileptic spikes and the observed decoupling. Although efforts were made to exclude slow oscillations and spindles temporally linked to epileptic spikes, it is possible that epileptic spikes originating from deeper structures (such as the hippocampus in mesial temporal lobe epilepsy) were not fully captured in scalp EEG recordings.<span><sup>3, 4</sup></span> Thus, consistent with our findings, one plausible explanation is that the association between epileptic spikes and slow oscillation–spindle coupling may critically depend on whether these spikes effectively propagate from the epileptogenic area to thalamus. If epileptic spike rates are high but only limitedly reach the thalamus, the global decoupling effect could be diminished, possibly explaining the lack of correlation in Schiller et al.'s cohort. Supporting this interpretation, the observed correlations between spike rates and slow oscillation–spindle coupling rates in their study, although not statistically significant, were consistently negative in both N2 and N3 sleep.</p><p>Altogether, these results reinforce the hypothesis that diverse epilepsies may share a common neurophysiological mechanism involving disrupted normal coordination of slow oscillations and spindles, with potentially significant implications for overnight memory consolidation.<span><sup>5</sup></span> We look forward to future studies delineating the conditions under which epileptic spikes propagate to the thalamus, the extent of their downstream network effects, and therapeutic st","PeriodicalId":11768,"journal":{"name":"Epilepsia","volume":"66 7","pages":"2600-2601"},"PeriodicalIF":6.6,"publicationDate":"2025-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1111/epi.18482","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144149895","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Deep learning-based classification and segmentation of interictal epileptiform discharges using multichannel electroencephalography.","authors":"Yulin Sun, Min Guan, Xun Chen, Fengling Feng, Runnan He, Lian Huang, Xiaoguang Tong, Huan Zhou, Xiuyun Liu, Dong Ming","doi":"10.1111/epi.18463","DOIUrl":"https://doi.org/10.1111/epi.18463","url":null,"abstract":"<p><strong>Objective: </strong>This study was undertaken to develop a deep learning framework that can classify and segment interictal epileptiform discharges (IEDs) in multichannel electroencephalographic (EEG) recordings with high accuracy, preserving both spatial information and interchannel interactions.</p><p><strong>Methods: </strong>We proposed a novel deep learning framework, U-IEDNet, for detecting IEDs in multichannel EEG. The U-IEDNet framework employs convolutional layers and bidirectional gated recurrent units as a temporal encoder to extract temporal features from single-channel EEG, followed by the use of transformer networks as a spatial encoder to fuse multichannel features and extract interchannel interaction information. Transposed convolutional layers form a temporal decoder, creating a U-shaped architecture with the encoder. This upsamples features to estimate the probability of each EEG sampling point falling within the IED range, enabling segmentation of IEDs from background activity. Two datasets, a public database with 370 patient recordings and our own annotated database with 43 patient recordings, were used for model establishment and validation.</p><p><strong>Results: </strong>The results showed prominent advantage compared with other methods. U-IEDNet achieved a recall of .916, precision of .911, F1-score of .912, and false positive rate (FPR) of .030 on the public database. The classification performance in our own annotated database achieved a recall of .905, a precision of .902, an F1-score of .903, and an FPR of .072. The segmentation performance had a recall of .903, a precision of .916, and an F1-score of .909. Additionally, this study analyzes attention weights in the transformer network based on brain network theory to elucidate the spatial feature fusion process, enhancing the interpretability of the IED detection model.</p><p><strong>Significance: </strong>In this paper, we aim to present an artificial intelligence-based toolbox for IED detection, which may facilitate epilepsy diagnosis at the bedside in the future. U-IEDNet demonstrates great potential to improve the accuracy and efficiency of IED detection in multichannel EEG recordings.</p>","PeriodicalId":11768,"journal":{"name":"Epilepsia","volume":" ","pages":""},"PeriodicalIF":6.6,"publicationDate":"2025-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144136115","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Forecasting epileptic seizures with wearable devices: A hybrid short- and long-horizon pseudo-prospective approach.","authors":"Mona Nasseri, Rachel E Stirling, Pedro F Viana, Jie Cui, Ewan Nurse, Philippa J Karoly, Vaclav Kremen, Matthias Dümpelmann, Gregory A Worrell, Dean R Freestone, Mark P Richardson, Benjamin H Brinkmann","doi":"10.1111/epi.18466","DOIUrl":"10.1111/epi.18466","url":null,"abstract":"<p><strong>Objective: </strong>Seizure unpredictability can be debilitating and dangerous for people with epilepsy. Accurate seizure forecasters could improve quality of life for those with epilepsy but must be practical for long-term use. This study presents the first validation of a seizure-forecasting system using ultra-long-term, non-invasive wearable data.</p><p><strong>Methods: </strong>Eleven participants with epilepsy were recruited for continuous monitoring, capturing heart rate and step count via wrist-worn devices and seizures via electroencephalography (average recording duration of 337 days). Two hybrid models-combining machine learning and cycle-based methods-were proposed to forecast seizures at both short (minutes) and long (up to 44 days) horizons.</p><p><strong>Results: </strong>The Seizure Warning System (SWS), designed for forecasting near-term seizures, and the Seizure Risk System (SRS), designed for forecasting long-term risk, both outperformed traditional models. In addition, the SRS reduced high-risk time by 29% while increasing sensitivity by 11%.</p><p><strong>Significance: </strong>These improvements mark a significant advancement in making seizure forecasting more practical and effective.</p>","PeriodicalId":11768,"journal":{"name":"Epilepsia","volume":" ","pages":""},"PeriodicalIF":6.6,"publicationDate":"2025-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144135885","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Precision medicine in epilepsy: Clinicians' perspectives from an international qualitative study.","authors":"Matthias De Wachter, Anne Juul, Annelies Colliers, Berten Ceulemans, Sarah Weckhuysen, Anna C Jansen, Rikke S Møller","doi":"10.1111/epi.18480","DOIUrl":"https://doi.org/10.1111/epi.18480","url":null,"abstract":"<p><strong>Objective: </strong>Precision medicine (PM) is gaining increasing importance in the treatment of rare genetic epilepsies. However, its availability and implementation in clinical practice remain limited. This study explores the barriers and facilitators influencing the implementation of PM for people with epilepsy (PWE).</p><p><strong>Methods: </strong>Semistructured interviews were conducted with clinicians involved in epilepsy care for PWE across various global regions. Participants were purposefully selected based on geographical distribution and World Bank income classification. Framework analysis was used to identify key themes.</p><p><strong>Results: </strong>Sixteen clinicians from six different continents were interviewed. Five key themes emerged. The implementation of PM depends on (1) the personal perspective and experience of health care providers; (2) the attitude of PWE and their caregivers toward PM, and their interaction with clinicians; (3) continuous education of health care providers, formation of expert teams, and generation of robust evidence on PM; (4) multilevel collaboration including patient advocacy groups; and (5) a clear, consistent organizational approach and the development and implementation of standardized guidelines.</p><p><strong>Significance: </strong>Clinicians consider PM as transformative for the care for PWE and expect it to redefine standard practice in the near future. Insufficient knowledge is the primary barrier to PM implementation, irrespective of socioeconomic context, highlighting the need for its integration into basic medical training and residency programs. To address inequities, integrated care pathways and standardized guidelines for genetic testing and PM, and the establishment of local PM-focused expert teams are essential to support clinicians in making informed PM decisions.</p>","PeriodicalId":11768,"journal":{"name":"Epilepsia","volume":" ","pages":""},"PeriodicalIF":6.6,"publicationDate":"2025-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144135890","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Serial correlation between saliva and blood beta-hydroxybutyrate levels in children commencing the ketogenic diet for epilepsy.","authors":"Neha Kaul, Jing Duan, Dong Cui, Michael Erlichster, Zhibin Chen, Dovile Anderson, Jianxiong Chan, Ingrid E Scheffer, Efstratios Skafidas, Jianxiang Liao, Patrick Kwan","doi":"10.1111/epi.18465","DOIUrl":"https://doi.org/10.1111/epi.18465","url":null,"abstract":"<p><strong>Objectives: </strong>Accurate and user-friendly methods to measure beta-hydroxybutyrate (BHB) concentration are needed to guide the optimal use of ketogenic diet therapy (KDT). We aimed to determine the correlation between serum, capillary, and salivary BHB concentration, and to validate an electrochemical salivary BHB point-of-care test (POCT) in children commencing KDT for drug-resistant epilepsy.</p><p><strong>Methods: </strong>This was a single center, prospective cohort study. Children <18 years with drug-resistant epilepsy electively admitted to Shenzhen Children's Hospital to initiate KDT between January 1, 2020 and June 30, 2021, were included. Over the 7-day admission, we collected paired saliva and capillary blood samples twice a day and serum blood samples on the first and last days of admission from each participant. Salivary BHB was measured using liquid chromatography mass spectrometry (LCMS) and the POCT. Primary outcome was the correlation between serum and salivary BHB concentration measured using LCMS. Secondary outcomes were the correlation between both the capillary blood and salivary BHB concentration, measured by LCMS, and the POCT device.</p><p><strong>Results: </strong>Seventy-one serum and 334 capillary blood paired with salivary samples were collected from 42 children (median age 4.5 years, interquartile range 1 to 8 years, 45% female). Salivary BHB measured using LCMS strongly correlated with serum BHB (Spearman's ρ = 0.910) and capillary blood BHB concentrations (Spearman's ρ = 0.865). Salivary BHB concentration was 6% and 7% of serum and capillary blood BHB concentration, respectively. The POCT demonstrated excellent test-retest reliability when compared with LCMS (ICC(A,k) = 0.983, 95% confidence interval: 0.980-0.986). Salivary BHB concentration measured by the POCT showed good accuracy in predicting the capillary blood BHB concentration within the therapeutic range of 2-5 mM.</p><p><strong>Significance: </strong>Salivary BHB concentration strongly correlates with both serum and capillary blood BHB concentration. The POCT accurately measures salivary BHB concentration and provides a simple, user-friendly method to guide the use of KDT for children with drug-resistant epilepsy.</p>","PeriodicalId":11768,"journal":{"name":"Epilepsia","volume":" ","pages":""},"PeriodicalIF":6.6,"publicationDate":"2025-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144135878","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The human photosensitive epilepsy model for clinical proof-of-principle trials of novel antiseizure medications. 1. Use of the EEG in drug development and characteristics of the model.","authors":"Dorothée Kasteleijn-Nolst Trenité, Wolfgang Löscher","doi":"10.1111/epi.18468","DOIUrl":"https://doi.org/10.1111/epi.18468","url":null,"abstract":"<p><p>Clinical development of novel antiseizure medications (ASMs) would benefit from an early proof-of-principle (POP) model. The photosensitivity model, which uses the photoparoxysmal electroencephalography (EEG) response (PPR) as a surrogate for seizures, is currently the only human model that allows POP trials of investigational compounds after a single drug administration. Typically, trials in this model are performed as single-blinded, placebo-controlled Phase IIa POP studies, evaluating a range of doses in small groups of patients with epilepsy. Although most patients in such trials exhibit generalized epilepsies, photosensitivity also occurs in focal-onset epilepsies. In the first part of this review, we describe the use of epileptiform discharges in drug testing, historical development of the photosensitivity model, the genetics and pathophysiology underlying the photosensitive response in patients with epilepsy, clinical characteristics of the patients, and details on drug testing. In the second part of this review, the outcome of numerous drug trials will be described in detail, including a critical discussion of the limitations of the model. In the past 50 years, the original and later standardized photosensitivity model has shown to be an unbiased, accurate, inexpensive method determining the potential efficacy of a novel ASM before entering large add-on trials with chronic drug administration, irrespective of the type of epilepsy.</p>","PeriodicalId":11768,"journal":{"name":"Epilepsia","volume":" ","pages":""},"PeriodicalIF":6.6,"publicationDate":"2025-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144135907","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Familial SYNGAP1 variants define the boundaries of a complex neurodevelopmental disorder with epilepsy.","authors":"Alicia G Harrison, Jan H Magielski, Ian McSalley, Shiva Ganesan, Anna J Prentice, Kristin G Cunningham, Samuel R Pierce, Michael J Boland, Benjamin L Prosser, Ingo Helbig, Jillian L McKee","doi":"10.1111/epi.18469","DOIUrl":"10.1111/epi.18469","url":null,"abstract":"<p><strong>Objective: </strong>SYNGAP1-related disorders are common neurodevelopmental conditions characterized by autism spectrum disorder, developmental delay, intellectual disability, and a range of generalized seizure types. Disease-causing variants in SYNGAP1 typically occur de novo. This study aims to characterize inherited cases of SYNGAP1-related disorders.</p><p><strong>Methods: </strong>Here we report three families including eight total individuals with inherited, protein-truncating variants in SYNGAP1, recruited through a natural history study. In two of the families, the proband inherited their pathogenic variant from a heterozygous parent. In the remaining family, the variant was inherited from a mosaic parent. This study additionally reports two families with inherited missense variants classified as variants of uncertain significance, which are not clearly diagnostic at this time.</p><p><strong>Results: </strong>Phenotypes in affected children and parents included both typical and attenuated SYNGAP1 presentations, including a single individual with a mosaic SYNGAP1 variant who was clinically unaffected. Among the individuals with protein-truncating variants, generalized epilepsy was observed in six individuals, autism spectrum disorder in two individuals, and developmental delay or intellectual disability in all individuals with germline variants.</p><p><strong>Significance: </strong>We demonstrate that SYNGAP1-related disorder can occur in families and that clinical presentations of familial cases are not limited to milder phenotypes. We estimate that 3% of cases of SYNGAP1-related disorder are inherited. Recognition of familial SYNGAP1-related disorders delineates edge cases of a relatively common neurodevelopmental disorder and has implications for variant interpretation and clinical practice.</p>","PeriodicalId":11768,"journal":{"name":"Epilepsia","volume":" ","pages":""},"PeriodicalIF":6.6,"publicationDate":"2025-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144126995","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Optimizing external validation of deep neural networks for interictal discharge detection","authors":"Galia V. Anguelova","doi":"10.1111/epi.18412","DOIUrl":"10.1111/epi.18412","url":null,"abstract":"","PeriodicalId":11768,"journal":{"name":"Epilepsia","volume":"66 7","pages":"2596-2597"},"PeriodicalIF":6.6,"publicationDate":"2025-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144126996","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Cortico-cortical evoked potentials: Analytical techniques and emerging paradigms for epileptogenic zone localization.","authors":"Zekai Qiang, Jamie Norris, Gerald Cooray, Richard Rosch, Kai Miller, Dora Hermes, Aswin Chari, Martin Tisdall","doi":"10.1111/epi.18467","DOIUrl":"https://doi.org/10.1111/epi.18467","url":null,"abstract":"<p><p>Cortico-cortical evoked potentials (CCEPs) are an active electrophysiological technique used during intracranial electroencephalography to evaluate the effective connectivity and influence of therapeutic stimulation between distinct cortical regions and pinpoint epileptogenic zones (EZs) in patients with epilepsy. Various methodologies have been implemented to analyze CCEPs and characterize the epileptogenic networks for EZ localization. Despite its promise, their interpretation remains challenging due to the large volumes of spatially and temporally complex data generated. Early studies focused largely on qualitative descriptors and predefined, semi-quantitative features such as waveform morphology and peak latencies. However, these methods are limited by the significant heterogeneity in CCEP waveform conformations across patients and cortical regions. The specific technique used for extraction of features, such as the spectral band power and root mean squared values, remains open to empirical refinement, as does choice of appropriate latency windows, with no consensus reached regarding the optimal approach. Graph theoretical metrics such as degree centrality, betweenness centrality, and clustering coefficients can provide a rich representation of epileptogenic network connectivity. However, these metrics are often abstract and difficult to interpret in a clinical setting or to the non-expert, and their neuroscientific substrates remains poorly understood. The lack of standardization in stimulation protocol and data-processing pipelines has further contributed to inconsistency in reported findings. Emerging machine learning approaches have been increasingly applied to CCEP data, offering a more data-driven and potentially generalizable way to identify electrophysiological biomarkers of the epileptogenic effective connectivity. In this article, we discuss qualitative, quantitative, and spectral features; network-analytical metrics; and more recently, data driven methodologies aimed at improving the interpretability and clinical utility of CCEP data.</p>","PeriodicalId":11768,"journal":{"name":"Epilepsia","volume":" ","pages":""},"PeriodicalIF":6.6,"publicationDate":"2025-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144126994","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Reply to: Optimizing external validation of deep neural networks for interictal discharge detection","authors":"Marleen C. Tjepkema-Cloostermans,&nbsp;Michel J. A. M. van Putten","doi":"10.1111/epi.18411","DOIUrl":"10.1111/epi.18411","url":null,"abstract":"&lt;p&gt;We appreciate the opportunity to respond to the letter regarding our study.&lt;span&gt;&lt;sup&gt;1&lt;/sup&gt;&lt;/span&gt; Although we welcome constructive discussion, the concerns raised reflect a misunderstanding of our methodology and the principles of external validation.&lt;/p&gt;&lt;p&gt;First, external validation assesses a model's generalizability in an independent clinical setting, and our study followed best practices by evaluating the deep learning model on an external dataset. The suggestion that our process introduced confirmation bias is misleading. Expert review of artificial intelligence (AI)-detected events is a widely accepted practice in clinical AI validation, particularly in the absence of a universally accepted gold standard for interictal epileptiform discharge (IED) detection.&lt;span&gt;&lt;sup&gt;2&lt;/sup&gt;&lt;/span&gt; Our inclusion of a multiexpert panel further strengthens the validation process and mitigates individual bias.&lt;/p&gt;&lt;p&gt;Second, although one of the original model developers participated in reviewing IEDs flagged by the model, final adjudication was conducted by a panel of five experts. This is a standard and accepted approach in electroencephalographic (EEG) studies, where interrater agreement naturally varies.&lt;span&gt;&lt;sup&gt;3&lt;/sup&gt;&lt;/span&gt; The assertion that this process introduces bias disregards that clinical neurophysiology often relies on expert consensus in the absence of a definitive ground truth.&lt;/p&gt;&lt;p&gt;Third, the claim that two authors who achieved perfect agreement (Cohen &lt;i&gt;κ&lt;/i&gt; = 1.0) were involved in both training and external validation, indicating a lack of “assessor independence,” misrepresents our study. The interrater variability in the internal validation set ranged from .71 to 1. The one pair of experts who achieved perfect agreement were from different institutions, each with &gt;20 years of experience in EEG interpretation. Their high &lt;i&gt;κ&lt;/i&gt; value reflects expertise, not a lack of independence. Furthermore, only one of them was involved in data labeling for training, internal validation, and the external validation panel, whereas the other was involved solely in internal validation.&lt;/p&gt;&lt;p&gt;Fourth, the suggestion that our validation process is susceptible to overfitting seems to stem from an essential misunderstanding of the concept, as overfitting pertains to the training phase rather than validation. Overfitting is a phenomenon occurring during training when a model is excessively tailored to a specific dataset, reducing its ability to generalize. External validation, by definition, does not involve retraining, making such concerns both misplaced and irrelevant.&lt;span&gt;&lt;sup&gt;4&lt;/sup&gt;&lt;/span&gt;&lt;/p&gt;&lt;p&gt;Finally, speculation about potential commercial influence is both unsubstantiated and misleading. Our study was conducted independently, with all affiliations transparently disclosed. Although commercial entities contribute to AI development in medicine, this does not inherently compromise scientific integrity when conflicts of interest are properly m","PeriodicalId":11768,"journal":{"name":"Epilepsia","volume":"66 7","pages":"2598-2599"},"PeriodicalIF":6.6,"publicationDate":"2025-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1111/epi.18411","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144126997","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}