Epilepsy Research

ImportanceImplantable sub-scalp EEG systems with a small number of channels have emerged as promising solutions for long-term seizure monitoring in patients with epilepsy. How seizure detection performance varies by montage configuration is unknown. ObjectiveTo quantify how automated seizure detection performance differs between full and reduced montages, and how these differences vary by epilepsy characteristics. DesignRetrospective cross-sectional study. SettingSingle-center at the Hospital of the University of Pennsylvania Epilepsy Monitoring Unit (EMU). ParticipantsEEG data from 2281 consecutive EMU admissions between January 2017 and December 2024 were screened. Admissions with at least one annotated seizure and one interictal clip [≥]20 minutes from any seizure were included. ExposureComputational simulation of published sub-scalp device montages using standard 10-20 EEG channels. Main Outcomes and MeasuresThe primary outcome was event-based F1 scores evaluated for three published seizure detectors--a one-class support vector machine (SVM), a convolutional neural network (SPaRCNet), and a long short-term memory autoregressive model (NDD)--across montages. ResultsA total of 466 admissions from 436 patients (mean [SD] age, 39.0 [14.4] years; 54.4% female) met inclusion criteria, comprising 1683 seizures and 1527 interictal clips. SPaRCNet achieved the highest performance (mean [SD] F1, 0.61 [0.30]), followed by NDD (0.56 [0.28]) and SVM (0.39 [0.25]). Performance decreased by at most 0.09 with reduced montages, depending on detectors. Patient factors accounted for the largest proportion of performance variance (29.2%), followed by detector choice (10.3%). Montage effects were minimal (0.4%), despite variation in optimal montage across detectors. Reduced-montage performance correlated moderately to highly with full-montage performance ({rho}=0.29-0.73), suggesting full-montage performance could help identify patients suitable for sub-scalp devices. Missed seizures were associated with lower amplitude and bandpowers than detected seizures, though they remained distinguishable from interictal data. Conclusions and RelevanceAutomated seizure detection achieved comparable accuracy, with only modest reductions, under simulated reduced montages. Performance differences were driven primarily by detector- and patient-level factors rather than montage. These findings support the feasibility of accurately detecting seizures with published sub-scalp devices and highlight the need for improved algorithms to optimize performance. Key FindingsO_ST_ABSQuestionC_ST_ABSHow do automated seizure detection algorithms perform with reduced-channel montages simulating published sub-scalp devices? FindingsIn this retrospective cross-sectional study, seizure detection performance decreased only modestly on reduced montages relative to the full montage (absolute F1 change -0.09 to 0.014), whereas patient- and algorithm-level factors accounted for most of performance variance (29.2% and 10.3%, respectively). Algorithm performance on full montage recordings was moderately correlated with performance on reduced channel montages ({rho}=0.29-0.73). MeaningReduced-montage sub-scalp devices are promising for ultra-long-term monitoring, but best performance requires selecting the right patients. Patient-specific seizure detectors will likely be required to optimize long-term performance.

Temporal lobe epilepsy (TLE) is a heterogeneous disorder with most clinical presentations involving unilateral or bilateral hippocampal seizure onsets. Antiseizure medications are often ineffective for TLE, and epilepsy surgery can have variable outcomes. Risk factors for TLE are readily identifiable and typically precede chronic epilepsy, providing a window of opportunity for preventative treatments. However, there are currently no clinically approved anti-epileptogenic therapies. In this study, we investigate the role of Wnt signaling in epileptogenesis using two mouse TLE models, the intrahippocampal kainate model of unilateral TLE (IHK), and the intraperitoneal kainate model of bilateral TLE (IPK). We specifically examined adult-born immature dentate granule cells as these cells have been heavily implicated in the pathogenesis of TLE and clinical TLE is typically initiated in adulthood. We observed that adult-born immature dentate granule cells undergo pathological morphological changes during epileptogenesis in both the IHK and IPK models of TLE. When compared across epileptogenic zones, however, these changes differed between the two models. Wnt signaling also decreased in these cells in epileptic mice during the epileptogenic period. When mice were treated with SB415286, a highly selective Wnt activator, Wnt signaling in immature dentate granule cells was restored to baseline levels and pathological remodeling changes were reduced in both models. These data therefore suggest that a reduction in Wnt signaling in immature dentate granule cells plays an etiological role in epileptogenesis, and that restoring Wnt signaling using Wnt activating drugs or alternative agents may have therapeutic potential as an anti-epileptogenic strategy in TLE.

Disease variants in GABR genes encoding {gamma}-Aminobutyric acid type A receptor (GABAAR) subunits are major causes of developmental and epileptic encephalopathies (DEEs). There is no effective treatment for these DEEs although the GABAAR is a major target for antiseizure drugs. We previously identified the therapeutic effect of 4-phenyl-butyrate (PBA) in Gabrg2+/Q390X knockin DEE mice and now test the effect of the drug in GABRA1 variants that encode the 1 subunit. We used a multidisciplinary approach including in silico structural modeling, flow cytometry, patch clamp recordings and bio-chemistry in conjunction with differential tagging of the wild-type and the mutant alleles to evaluate the effect of PBA on rescue of GABAAR subunit expression, surface trafficking, and function in vitro in heterologous HEK293T cell model and in vivo in Gabra1+/A322D mice. We found that both total and cell surface 1 expression was reduced when the variant 1 protein was present; suggesting reduced functional receptor on the cell membrane and synapse. Patch clamp recordings identified 1 variants reduced GABA-evoked current amplitude. In silico prediction indicated reduced protein stability for GABRA1 variants indicated by negative {Delta}{Delta}G values. PBA increased both total and surface expression of wildtype 1 and 1 variants; and improved expression of both wildtype and variant 1 alleles when these were co-expressed. Importantly, PBA also increased the GABAAR expression in the thalamus of the Gabra1+/A322D mice. This study indicates that PBA is a promising treatment option for DEEs associated with GABRA1 mutations. Our previous work has demonstrated that PBA improves proteostasis by enhancing expression of the wildtype allele, repairing the mutant allele, and reducing endoplasmic reticulum stress. Therefore, it can mitigate seizures and improve neurobehavioral phenotypes at behavioral levels. Based on this and our previous work on GABRG2 and SLC6A1 mutations, we propose that PBA holds promise as a common medicine for multiple genetic neurologic disorders that share the proteostasis pathology with a broad clinical application in DEEs.

BackgroundIntegrating multimodal data into medical artificial intelligence (AI) tools and evaluating whether they outperform human experts remains a critical challenge. Epilepsy surgery offers a unique paradigm for this evaluation, as it provides an expert-independent measure (Engel score) of post-surgical outcome. Currently, evaluation for epilepsy surgery relies on the visual interpretation and human synthesis of multimodal data. While clinical evaluations are individualized and account for complex anatomical variability, integrating these diverse, high-dimensional modalities to generate a probability of surgical success remains challenging. Here, we leverage this objective outcome score to investigate the feasibility of a data-driven, phenotype-based model against the current clinical gold standard. MethodsThe evaluation was performed on an epilepsy-type controlled cohort of 57 patients from six tertiary epilepsy surgery centers who underwent resective/ablative surgery in the mesiotemporal lobe. Multimodal data, namely, patient demographics, semiology, invasive electrophysiology monitoring, and neuroimaging, were utilized. We first estimated how human experts perceive surgery success. Subsequently, we developed a data-driven model integrating these modalities to predict surgery outcomes. The model performance was compared to the current clinical gold standard (three independent human experts) and published outcome calculators. Finally, modality-level phenotypes were derived based on the models predictions. ResultsPredictions by human experts correlated poorly with post-surgical outcomes, and published outcome calculators did not perform better than the experts (DeLongs p = 0.367). Our model incorporating multimodal data achieved an area under the receiver operating characteristic curve (AUROC) of 0.801. It performed statistically better than the best human expert (DeLongs p = 0.043) and achieved a higher AUROC than the best published surgical outcome calculator (0.801 vs. 0.694). ConclusionsWe demonstrated the proof-of-concept that data-driven multimodal phenotypes can inform personalized surgery planning in epilepsy. Furthermore, we provide a framework for integrating multimodal data and benchmarking medical AI performance against human experts.

AIM: STXBP1-related disorder (STXBP1-RD) is a severe developmental and epileptic encephalopathy characterized by early-onset seizures and persistent cognitive and motor impairments. With disease-modifying trials emerging, a disorder-specific severity scale is needed. To address this, we adapted a validated clinician-reported measure from CDKL5 Deficiency Disorder to develop the STXBP1 Clinical Severity Assessment (S-CSA) and evaluated its psychometric properties. METHOD: The S-CSA was adapted from the CDKL5 Clinical Severity Assessment through expert consensus sessions with STXBP1 clinicians. Revisions addressed gaps in motor and vision domains, adding tremor and vision items. The measure was administered to 123 individuals with STXBP1-RD. Psychometric evaluation included confirmatory factor analysis, internal consistency, composite reliability, average variance extracted, and distinctiveness, compared with recommended thresholds. RESULTS: Analyses supported a three-domain structure (motor, communication, vision) with factor loadings >0.5 and strong internal consistency (Cronbachs alpha >0.7; composite reliability >0.88). Model fit and variance metrics met recommended standards, and domains demonstrated distinctiveness. No ceiling or floor effects were observed. Minimal skew was seen in motor (0.34) and communication (0.16) domains; positive skew in vision (2.2) was seen, identifying patients with and without cortical visual impairment. INTERPRETATION: The S-CSA demonstrates strong validity and reliability in STXBP1-RD and may show utility in clinical trials for STXBP1-RD and potentially other severe DEEs. Key Words: STXBP1-Related Disorder, Developmental and Epileptic Encephalopathies, Clinical Outcome Assessments

IntroductionTo leverage high-frequency oscillations (HFOs) as a biomarker with significant potential, this study compared a large set of detectors on a unified dataset, aiming to evaluate their clinical applicability under realistic conditions. MethodsEleven automatic detectors were applied to a retrospective dataset of intracranial and scalp EEGs from 27 consecutive pediatric patients. Inter-detector agreement was assessed using Spearmans Rho, and the area under the curve (AUC) for seizure onset zone (SOZ) prediction served as a consistent reference standard to enable reliable comparisons across recording modalities. Analyses were conducted separately for HFO and Spike-HFO detections. ResultsThe average age of our cohort was 12.4 years (SD 4.0; range 5-18). AUC values in scalp EEG ranged from 0.61 to 0.67 for HFOs and from 0.53 to 0.63 for Spike-HFO. AUC values in intracranial EEG ranged from 0.48 to 0.66 for HFOs and 0.54 to 0.69 in Spike-HFO. Although only three of the 11 detectors were specifically developed or adapted for scalp EEG, the detectors generally achieved higher AUC values and stronger agreement in scalp EEG ConclusionsWe present the first study comparing intracranial and scalp detectors by testing them beyond the modalities for which they were originally designed. Although the clinical utility of detections was comparable across EEG modalities, it remained lower than reported in original studies assessing the diagnostic value of HFOs. Caution is warranted when applying a publicly available detector to a new dataset, and detector robustness remains a critical issue. Key points- A comprehensive head-to-head comparison of 11 detectors demonstrated significant variability in detector agreement and clinical utility - Clinical utility was not necessarily linked to the EEG recording type the detector was originally designed for - Despite widely accepted use of automatic detections, detector robustness remains a critical issue

Objective Memory impairment is a frequent comorbidity of focal epilepsy, incompletely explained by seizure frequency or structural pathology. Ictal and postictal hippocampal dysfunction disrupt memory processes, but their cumulative impact remains poorly quantified. This study introduces cumulative hippocampal seizure-related burden metrics and examines their association with long-term memory consolidation. Methods Twenty consecutive patients undergoing stereo-EEG in Marseille (2016-2018) were prospectively included. Continuous stereo-EEG recordings between two memory assessments (30 minutes and one week post-encoding) were analysed. Hippocampal ictal involvement and durations were assessed using epileptogenicity markers and visual stereo-EEG analysis. The postictal period was quantified using permutation entropy. Cumulative hippocampal seizure-related burden metrics (ictal, postictal and combined: c-HipSZB) were computed across hippocampus-involving ictal events. Verbal and visual memory were assessed using standardized recall and recognition tasks. Associations were examined using univariate and multivariate analyses. Results Higher dominant-hemisphere hippocampal burden was associated with poorer one-week verbal memory (performance and retention), independently of most covariates. Higher c-HipSZB was associated with lower total recall performance (RT; free + cued) and RT retention ({beta} = -25.04 and -23.88; R2 = 0.57 and 0.53; p < 0.05) and accounted for the greatest variance in both outcomes (adjusted R2= 0.59 and 0.53; {beta} = -25.45 and -24.27; p < 0.01), particularly when adjusting for epilepsy duration. No robust associations were observed between non-dominant-hemisphere hippocampal seizure-related burden metrics and visual memory. Effects predominantly involved recall. Interpretation Cumulative ictal-postictal hippocampal dysfunction is a major determinant of impaired long-term verbal memory consolidation in focal epilepsy.

Background and Purpose: Drug resistant epilepsy (DRE) affects approximately 15 million people worldwide, and surgery remains the only curative option. A key challenge in predicting outcomes is the lack of standardized, quantitative tools to help distinguish seizure driver regions from responder regions during stereoelectroencephalography (sEEG) recordings. We validated the CN Suite, a computational platform that uses causal network mapping and machine learning to assign criticality scores to sEEG contacts, testing whether higher scores correspond to surgically treated tissue in patients with favorable outcomes. Methods: We analyzed deidentified clinical data from 60 patients (aged 2 years and older) with focal or multifocal DRE who underwent sEEG monitoring and proceed to surgery at four U.S. Level 4 epilepsy centers. The algorithm was trained on an independent cohort (N=37) and locked prior to validation. The primary outcome was the standardized effect size (Cohens d) of the patient level surgical zone enrichment ratio between more favorable (Engel I or II) and less favorable (Engel III or IV) outcome groups. Contact level sensitivity, specificity, PPV, and NPV were evaluated at a prelocked threshold. Results: The findings support our hypothesis: the algorithm results showed significantly higher criticality values for surgically treated tissue in favorable outcome patients (d=0.74, 95% CI: 0.39 to 1.06, p=0.003). Three potentially clinically actionable findings emerged. First, high-criticality contacts formed spatially compact clusters (~9 mm nearest-neighbor distance vs. 17mm expected by chance), consistent with focal targets amenable to minimally invasive ablation. Second, sensitivity was highest in small focal procedures (80% at 10 or fewer treated contacts) and decreased with resection size. Third, in patients whose surgery failed, high-critical tissue remained outside the resection boundary, suggesting incomplete treatment coverage of the epileptogenic zone rather than mislocalization. Prediction specificity was 84% at the contact level. For adult and pediatric cases (n=28), 88% of contacts that were identified as seizure free were in fact seizure free. Conclusions: Causal network mapping of sEEG identifies compact epileptogenic targets that correspond to surgically treated tissue in patients with more favorable outcomes. CN-Suite performed best in focal procedures and may be best suited for LITT and other minimally invasive approaches. In addition, low-criticality regions were infrequently associated with seizure-generating tissue, particularly in the pediatric cohort although our sample size was small. When surgery failed, residual high-critical tissue outside the resection boundary offered both a mechanistic explanation for less favorable surgical outcome as well as a potential target for reoperation.

Prader-Willi syndrome (PWS) is a rare multisystemic disorder characterized by obesity, endocrine dysfunctions, and psychiatric comorbidities, which imply frequent use of psychotropic medications. They account for atypical responses to standard dosages of psychiatric drugs. Pharmacogenetics could be part of the reason for this situation, potentially offering a valuable tool for individualized treatment. This study analyzed allelic and phenotypic frequency distributions of five of the main cytochrome P450 enzymes (CYP2D6, CYP2B6, CYP2C19, CYP2C9, CYP3A4) involved in psychiatric drug metabolism in 47 patients with genetically confirmed diagnosis of PWS and compared them to reference frequencies in the general European population. Allelic frequency comparisons between the European reference population and the overall PWS cohort revealed a significant global difference for CYP2B6, with CYP2C19 and CYP2D6 showing trends toward significance. Although no global allelic differences remained significant after false discovery rate correction, post-hoc analyses consistently identified an enrichment of reduced- or non-functional alleles CYP2B619 and CYP2D610 in patients with PWS. Predicted metabolizer phenotype analyses showed a significant shift toward intermediate metabolizers of CYP3A4 in the PWS cohort, with corresponding depletion of normal metabolizers. Subgroup analyses indicated that allelic differences were more pronounced in maternal uniparental disomy and non-deletion subtypes, particularly for CYP2B6, although no significant differences were observed between PWS genetic subtypes. Overall, results imply potential differences in metabolizing activity in PWS patients, and subsequent implications in drug efficacy and tolerability. These results support the idea that pharmacogenetic testing may improve therapeutic decision-making in PWS for psychiatric treatment. Larger studies are needed to confirm these preliminary results.

Stress is a commonly reported seizure precipitant and may contribute to the development of psychiatric comorbidities in epilepsy, yet how chronic stress interacts with epileptic circuits remains poorly understood. We investigated the impact of chronic restraint stress on physiological, behavioral, and synaptic outcomes in a mouse model of Dravet syndrome, specifically corticotropin-releasing factor (CRF) neurons in the bed nucleus of the stria terminalis (BNST), a stress-responsive region implicated in epilepsy patients. Chronic restraint stress produced divergent hypothalamic-pituitary-adrenal axis responses, with stressed Dravet syndrome mice exhibiting elevated corticosterone, increased mortality in females, and increased locomotion and anxiety-like behavior. Ex vivo electrophysiological recordings revealed that chronic stress increased spontaneous excitatory event frequency onto BNST CRF neurons in both genotypes and selectively increased sEPSC and sIPSC amplitude in Dravet syndrome mice. Evoked recordings demonstrated genotype-specific effects of stress on glutamatergic transmission in CRF neurons of the DS group. This suggests greater stress-dependent remodeling of spontaneous and evoked synaptic activity in DS. These findings suggest chronic stress may worsen physiological and behavioral outcomes in Dravet syndrome and promote specific maladaptive alterations in BNST CRF circuitry. More broadly, these results suggest that stress interacts with seizure vulnerability and potentially contributes to neuropsychiatric comorbidities and epilepsy.

Purinergic 2X7 receptor (P2X7R) is considered to play a critical role in neurological diseases, including epilepsy, and has also been proposed as a potential marker for neuroinflammation. This study aimed to validate the binding properties of the novel P2X7R radiotracer, [3H]JNJ-64413739, in rat brain using in vitro autoradiography, and additionally to explore spatial and temporal changes in P2X7R binding levels in a rat model of temporal lobe epilepsy using intrahippocampal administration of kainic acid (KA). Saturation of [3H]JNJ-64413739 to brain sections yielded a KD of approximately 3 nM, with full saturation around 10 nM. The radiotracer was displaced with a structurally different P2X7R ligand, JNJ-47965567, indicating high affinity and specificity to rat P2X7R. In post epileptic rats, region-specific [3H]JNJ-64413739 binding revealed a bilateral increase in the hippocampal formation and its subregions few days after status epilepticus, peaking at day 30, and remained stable at this high level until day 90. Similar temporal profiles were identified in subcortical regions such as the thalamus. Interestingly, no change in binding was observed in the temporal and piriform cortices until day 30 where a dramatic increase occurred. Also, in the corpus callosum, significant increase was detected 30 days after the seizure. These results show that P2X7R binding, likely reflecting inflammation, is increased at delayed time points and exhibit region-specific patterns that is different from acute effects. Our findings suggest that P2X7R may contribute to sustained neuroinflammation and may be involved in those changes leading to epileptogenesis and the development of chronic epilepsy. Highlights[3H]JNJ-64413739 binds specifically to the purinergic P2X7 receptor (P2X7R) and saturates in the rat brain. P2X7R binding increases in a region- and time-dependent manner following status epilepticus. P2X7R binding remains elevated during chronic epilepsy in all examined brain regions. P2X7R is considered a link between early seizures and sustained neuroinflammation and epileptogenesis.

Objective: Interictal spikes have been proposed as a biomarker for both localizing seizure onset zones (SOZ) and tracking changes in seizure risk with neurostimulation in patients with drug-resistant epilepsy. Electrical stimulation can modulate spike rates acutely, and it has been proposed that measuring this modulation can help localize the SOZ. However, it is unclear whether stimulation-induced spike rate changes reflect epilepsy-specific pathology in the stimulated network or simply intrinsic regional excitability, which limits our understanding of their utility in epilepsy surgery planning. Methods: We analyzed low-frequency stimulation (LFS; 1 Hz) applied during a clinical seizure-induction protocol systematically targeting multiple brain regions in 43 patients with drug-resistant epilepsy undergoing intracranial EEG monitoring. A validated, automated spike detector was used to quantify pre-, during-, and post-stimulation spike rates. We tested whether the stimulation-evoked spike rate response (i) tracks the expected change in seizure risk from a seizure induction protocol, (ii) varies with anatomical stimulation site and epilepsy localization, (iii) localizes the SOZ beyond baseline spike rate, and (iv) is accompanied by changes in spike morphology. Results: Nearby LFS acutely increased spike rates in high-spiking channels (inter-stimulation median 2.25 vs. during-stimulation 4.25 spikes/min; p < 0.001), with effects attenuating with distance and resolving within approximately 30 seconds of stimulation offset. Mesial temporal lobe stimulation produced the largest increase in nearby spike rates relative to temporal neocortex and other cortex (Kruskal-Wallis p = 0.003), but this effect did not differ between patients with and without mesial temporal lobe epilepsy. A random forest classifier incorporating stimulation-evoked modulation features achieved an AUC of 0.787, comparable to a resting-state spike model (AUC 0.747; DeLong p = 0.81), indicating that stimulation-evoked spike changes do not add localizing information beyond resting-state spike rates. Stimulation produced a small but significant shift in spike morphology toward broader, higher-amplitude discharges (PERMANOVA p < 0.001), consistent with recruitment of a broader neuronal population. Significance: LFS-evoked increases in interictal spike rates reflect intrinsic regional excitability, greatest in the mesial temporal lobe, rather than epilepsy-specific pathology, and do not improve SOZ localization over resting-state spike rates. These results argue against using the change in spikes with stimulation to localize the SOZ. On the other hand, the transient spike rate increase induced by a pro-epileptic protocol supports the acute change in spike rate as a biomarker of the effect of stimulation on seizure risk, with potential to guide parameter selection for epilepsy neuromodulation.

Background and ObjectivesFunctional neurological disorder (FND) is one of the most common causes of neurological symptoms and disability, but much remains unknown about its pathophysiology. In both clinical conversations and research publications, clinicians and researchers imply a variety of models for onset of the condition with respect to both the process culminating in its onset, and the distribution of susceptibility to the condition across the population. Here we used population-level data as evidence to arbitrate between these generative models of the condition. MethodsWe identified six hazard distributions corresponding to different pathophysiological processes, and four distributions of population susceptibility, as the assumptions underlying the range of plausible generative models resulting in the observed distribution of age of onset of FND. We combined these model families into 24 parametric proportional hazards models, and fitted each to the observed distribution of reported age at onset in two large FND datasets, one for functional movement disorders (FMD) and one for functional seizures (FS). Out-of-sample predictive accuracy for these models was compared using Bayesian model comparison. ResultsStrong trends were seen across model families with different distributions of population susceptibility to FND. For both datasets, the best-fitting model family overall was the mixture-cure family, which represents susceptibility as binary, with a susceptible and an unsusceptible proportion of the population. For the FMD dataset, some models in the log-normal frailty family had comparable fits to the mixture-cure models, and for the FS dataset, a number of the gamma frailty family had comparable fits. The variance parameters for each of these frailty distributions were so large as to imply binary risk, approximating mixture-cure models. Models with exponential hazard distributions--which correspond to a generative process where a single trigger in a susceptible person brings about the condition--were universally poor fits for the observed data. Other hazard distributions were insufficiently distinguished by their out-of-sample predictive accuracy to make further inference as to the underlying process resulting in onset of FND in susceptible individuals. InterpretationOur results suggest that susceptibility to FND is approximately binary, with the susceptible proportion of the population extremely likely to develop FND in their lifetime. The results also argue strongly against a generative model where a single trigger is sufficient to cause the onset of FND in a susceptible person.

ObjectiveThe ketogenic diet (KD) is a high-fat, low-carbohydrate intervention widely used to treat drug-resistant epilepsy, thought to reduce seizures through a combination of metabolic, neuronal, and microbiota-dependent mechanisms. Additionally, recent studies suggest that the anticonvulsant effects of KD require the gut microbiota, with taxa such as Akkermansia and Parabacteroides contributing to seizure protection by modulating host neurotransmitter balance and neural excitability. While KD has been shown to be effective in reducing seizure burden across different epilepsies, its antiseizure effect on infection-driven seizures, which are often driven by acute neuroinflammation, has not been evaluated. Here, we evaluated the effects of KD on seizure burden, neuroimmune responses, and gut microbiota composition in the Theilers murine encephalomyelitis virus (TMEV) model of virus-induced epilepsy. MethodsMice were maintained on either a KD or a normal diet prior to intracerebral TMEV infection. Seizures were induced by handling and scored twice daily from day 3 to 7 post-infection. Neuroimmune responses were assessed by flow cytometry, and fecal microbial composition was analyzed using 16S rRNA gene sequencing. ResultsDespite achieving ketosis, KD did not reduce seizure incidence, seizure burden, or seizure severity during acute TMEV infection. KD also did not significantly alter overall immune cell infiltration into the central nervous system, indicating limited effects on global neuroinflammation. However, KD significantly reshaped the gut microbiota, reducing alpha diversity (richness, Shannon diversity, and evenness) and strongly altering community structure with clear separation between diet groups, including enrichment of taxa such as Akkermansia, Acetatifactor, Dorea, and Flintibacter, and depletion of fiber-associated taxa including Bifidobacterium and Roseburia. However, these microbial shifts were insufficient to mitigate inflammation-driven seizures. SignificanceThese results demonstrate that KDs anticonvulsant efficacy is highly context-dependent, and that KD-driven changes in microbiota- and metabolite-mediated mechanisms may be ineffective against infection-associated epilepsy, suggesting that inflammation-driven seizures require distinct therapeutic approaches. Key pointsO_LIThe ketogenic diet (KD) does not reduce acute seizure incidence and severity during TMEV infection despite achieving ketosis C_LIO_LIKD does not induce neuroinflammatory changes associated with seizure outcomes C_LIO_LIKD strongly reshapes gut microbiota, reducing diversity and altering community structure. C_LIO_LIMicrobiota changes are insufficient to protect against inflammation-driven seizures C_LIO_LIKD anticonvulsant effects are context-dependent and ineffective in infection-driven epilepsy C_LI

Gain-of-function variants (GOF) in SCN8A, which encodes the NaV1.6 sodium channel, lead to epilepsy syndromes ranging from drug-responsive self-limited (SeLIE) and intermediate epilepsy to drug-resistant developmental and epileptic encephalopathy (DEE). It is currently unclear why individuals with SCN8A GOF variants show variable responses to sodium channel blockers (SCBs). Here, we compared the clinical characteristics of 173 individuals with 25 different SCN8A GOF variants following the hypothesis that carriers of variants affecting activation gating respond less well to SCBs than those with variants affecting fast inactivation gating, given that use-dependent SCBs preferentially target inactivated channel states. We found that individuals with variants altering channel activation gating were more severely affected than those with variants altering inactivation properties: They had an earlier age at onset (3 vs. 5 months, P < 0.0001), higher prevalence of DEE (75% vs. 39%; P < 0.0001), and poorer response to SCBs (20% vs. 69% seizure free; P < 0.0001). We performed pharmacological studies on representative and recurrent variants from each group: two variants (F846S and M1760I) causing hyperpolarizing shifts of the voltage-dependent activation curves, and two variants (G1475R and N1877S) causing depolarizing shifts of the voltage-dependent fast inactivation curves. Phenytoin failed to suppress neuronal firing in neurons expressing activation-related variants, but showed good suppressing effects in neurons expressing inactivation-related variants. In contrast, PRAX-330, a new SCB, which showed much faster binding rates than phenytoin, was effective for both groups of variants by markedly reducing neuronal firing through rapidly and persistently stabilizing NaV1.6 in the inactivated state. Our findings provide new insights into the mechanism of drug-resistance in SCN8A-DEE and support PRAX-330 and compounds with similar pharmacological properties as a promising preclinical candidate for targeted therapies.

Background: Large language models (LLMs) demonstrate strong performance in controlled medical environments such as multiple choice exams, but their utility in real-world clinical workflows remains unproven. The NHS Advice & Guidance (A&G) service, where Primary Care clinicians can submit text-based queries to specialists, provides an environment for evaluating the clinical performance of LLMs as a specialist. Methods: We compared responses from MedGemma 4B-IT, an open-weight model deployed locally on hospital infrastructure, against specialist neurologist responses across 50 adult neurology A&G cases from University College London Hospital. Two neurologists and two GPs rated 80 blinded and 20 unblinded responses for outcome, safety, efficacy, and feasibility using standardised criteria; outcome was a binary correct/incorrect, while other domains were scored 1-5. Inter-rater reliability was assessed using intraclass correlation coefficients. Results: Although there were no statistically significant differences between blinded specialist neurologists and LLM responses across any domain (outcome: 84% vs 82%, p=0.67; safety: 3.98 vs 4.02, p=0.85; efficacy: 4.06 vs 3.98, p=0.61; feasibility: 4.39 vs 4.20, p=0.45), 10% of LLM responses received concerning scores ([&le;]2 average score) compared to 0% of human responses, indicating potentially clinically important tail risk. Furthermore, unblinded results showed a preference for human responses, with human ratings being preferred across all domains. Only 51% of binary outcomes had unanimous agreement and inter-rater agreement was moderate across other domains (ICC 0.50-0.52). Conclusions: In this pilot study, aggregate scores between blinded human and LLM responses were similar, and no statistically significant differences were detected in this exploratory sample. However, aggregate metrics masked clinically important edge-case failures in LLM responses. Pronounced inter-rater variability and the potential impact of LLM/human syntax on blinded rater judgements highlight the challenges in establishing robust evaluation frameworks for clinical LLM deployment

Objective: Cognitive impairment is common among older adults with epilepsy, although efficient screening tools suitable for routine use are lacking. Here we examine, for the first time, the utility of the Alzheimers Disease Assessment Scale-Cognitive Subscale (ADAS-Cog) as a screening tool to identify cognitive impairment in older adults with epilepsy. Methods: Participants included 83 adults (ages over 55) with epilepsy from the Brain, Aging, and Cognition in Epilepsy (BrACE) study and 83 age-, sex-, and education-matched cognitively healthy controls from the Alzheimers Disease Neuroimaging Initiative (ADNI-3). All completed the ADAS-Cog and a comprehensive neuropsychological battery to identify cognitive phenotypes (intact vs impaired). Performance on individual ADAS-Cog items and the total score was assessed, and diagnostic efficiency statistics were determined. Results: Epilepsy participants (mean age=66.4 years) performed significantly worse across the ADAS-Cog total score and 8 of the 13 individual test items compared to controls. The largest effect sizes were observed on verbal learning and memory tasks, particularly word recall (d=0.87) and delayed word recall (d=1.06). An ADAS-Cog total score of at or exceeding 15 yielded optimal diagnostic efficiency (67.5% accuracy, 68.8% sensitivity, 66.7% specificity) for identifying cognitive impairment. Significance: The ADAS-Cog is sensitive to detecting cognitive impairment in older adults with epilepsy and may represent a scalable screening option in this population. Additional comparative studies in older epilepsy populations are needed to determine the sensitivity of this measure to longitudinal change, cross-cultural applicability, and availability across languages. Plain language summary: Cognitive decline is common among older adults with epilepsy, although sufficient evidence supporting the use of screening tools to identify cognitive impairment in this population is lacking. The ADAS-Cog may be a useful screening option in epilepsy research and clinical care, although additional studies are needed to compare it with other cognitive screening tests and to confirm its applicability for clinical care and across cultures and healthcare settings.

Electroencephalography (EEG) interpretation in clinical practice relies on the analysis of energy distribution across standard frequency bands. The Weed Plot framework encodes band-wise spectral energy, computed using Fourier-based methods, into a symbolic representation that preserves the interpretability of traditional EEG analysis. In this study, we propose a wavelet-based extension of this framework, where the energy of predefined clinical EEG bands is estimated using the Discrete Wavelet Transform instead of Power Spectral Density. Unlike Fourier-based approaches, wavelets provide a time-frequency representation that captures transient and non-stationary dynamics while remaining consistent with clinically defined bands. From these estimates, symbolic patterns are constructed based on the relative ordering of frequency bands within short temporal windows. Their empirical distribution is used to extract entropy-based features for epilepsy detection using multiple machine learning classifiers. From an Artificial Intelligence perspective, the main contribution is a structured symbolic encoding that enhances feature discriminability. From an engineering perspective, the contribution lies in an automated framework for EEG-based epilepsy detection. Experimental results show that wavelet-based representations improve classification performance compared to raw entropy and Fourier-based features. This improvement arises from the interaction between time-frequency localization and symbolic encoding, producing more discriminative feature distributions. These findings support wavelet-based symbolic representations as a robust and interpretable framework for EEG analysis, bridging clinical interpretation and data-driven methods.

Objective: To determine whether absolute ictal energy on intracranial EEG identifies brain regions whose epileptogenic involvement is attenuated under existing baseline-normalized, dynamic-systems, and event-based frameworks. Approach: Intracranial EEG from 56 patients (five centers; 21 SEEG, 35 ECoG) was analyzed using the Teager-Kaiser Energy Operator computed as z-scored and raw envelopes; energy-dominant network regions (EDNRs) were defined as electrodes whose raw-energy rank exceeded their z-score rank by at least 2 positions. Hilbert decomposition characterized instantaneous amplitude and frequency. Main results: EDNRs were identified in 51 of 56 patients (91%; mean 3.4). Hilbert decomposition revealed elevated baseline amplitude in EDNRs relative to both non-involved regions (p < 0.001) and potential seizure onset zones (PSOZs, the top-ranked electrodes under both metrics; p = 0.029), with EDNRs participating in seizure-frequency dynamics comparable to PSOZs (mean ictal frequency shift +3.7 versus +4.1 Hz). EDNR detectability correlated directly with electrode count (Spearman r = 0.899, p < 0.001) without plateau. Significance: Absolute ictal energy identifies an epileptogenic network component with elevated baseline amplitude attenuated under baseline-normalized metrics. The dual-metric framework defines a complementary energy-based axis and establishes the second layer of a two-layer approach with seizure onset and propagation mapping as the first layer. EDNR detectability scales with electrode count, directly relevant to SEEG implantation strategy and to network-level inferences from heterogeneously covered cohorts.

BackgroundBipolar disorder (BD) is frequently underdiagnosed, particularly in patients presenting with depressive disorders, leading to delays in appropriate treatment. Artificial intelligence (AI) applied to electronic health records (EHRs) may improve early detection by identifying clinically relevant symptom patterns. ObjectiveTo evaluate the diagnostic performance of a natural language processing (NLP)-based AI model for detecting BD-related features in EHRs of patients with affective diagnoses. MethodsA retrospective diagnostic accuracy study was conducted using 500 EHRs from a psychiatric referral hospital in Bogota, Colombia (2020-2024). The model extracted 18 predefined clinical domains from unstructured text and classified patients into four risk categories. Diagnostic performance was assessed in a validation subset of 100 records using independent psychiatric evaluation as the reference standard. Sensitivity, specificity, positive and negative predictive values, F1-score, and area under the receiver operating characteristic curve (AUC-ROC) were calculated. ResultsThe model achieved high agreement in symptom extraction (mean 91.1%). Sensitivity was 96.4% (95% CI: 87.7%-99.0%) and specificity was 84.4% (95% CI: 71.2%-92.3%), with an F1-score of 0.92 and an AUC-ROC of 0.932 (95% CI: 0.881-0.975). A substantial proportion of patients with depressive diagnoses were identified as having confirmed BD or clinically relevant risk. The model analyzed complete EHRs 120 times faster than human reviewers. ConclusionsNLP-based analysis of EHRs can achieve clinically meaningful performance in identifying BD-related patterns while substantially reducing review time. The model may be useful as a clinical decision support tool for earlier identification of bipolar disorder.