Epileptic seizure focus detection from interictal electroencephalogram: a survey

Sleep can modulate epileptic activities, but our knowledge of sleep perturbation by epilepsy remains sparse. Interestingly, epilepsy and sleep both present with defining electrophysiological features in the form of specific graphoelements on EEG. This raises the possibility to identify, within ongoing EEG activity, how epilepsy impacts and disrupts sleep. Here, we asked whether the presence of a lateralized epileptic focus interferes with the expression of the dominant electrophysiological hallmarks of sleep: slow oscillations, slow waves and spindles. To this aim, we conducted a cross-sectional study and analysed sleep recordings with surface EEG from 69 patients with focal epilepsy (age range at EEG: 17–61 years, 29 females, 34 left focal epilepsy). Comparing patients with left and right focal epilepsy, we assessed inter-hemispheric asymmetry of sleep slow oscillations power (delta range, 0.5–4 Hz); sleep slow wave density; amplitude, duration and slope; and spindle density, amplitude, duration as well as locking to slow oscillations. We found significantly different asymmetries in slow oscillation power (P < 0.01); slow wave amplitude (P < 0.05) and slope (P < 0.01); and spindle density (P < 0.0001) and amplitude (P < 0.05). To confirm that these population-based differences reflect actual patient-by-patient differences, we then tested whether asymmetry of sleep features can classify laterality of the epileptic focus using a decision tree and a 5-fold cross-validation. We show that classification accuracy is above chance level (accuracy of 65%, standard deviation: 5%) and significantly outperforms a classification based on a randomization of epileptic lateralization (randomization data accuracy: 50%, standard deviation 7%, unpaired t-test, P < 0.0001). Importantly, we show that classification of epileptic lateralization by the canonical epileptic biomarker, i.e. interictal epileptiform discharges, improves slightly but significantly when combined with electrophysiological hallmarks of physiological sleep (from 75% to 77%, P < 0.0001, one-way ANOVA + Sidak’s multiple comparisons test). Together, we establish that epilepsy is associated with inter-hemispheric perturbation of sleep-related activities and provide an in-depth multi-dimensional profile of the main sleep electrophysiological signatures in a large cohort of patients with focal epilepsy. We provide converging evidence that the underlying epileptic process interacts with the expression of sleep markers, in addition to triggering well-known pathological activities, such as interictal epileptiform discharges.

Acute effects of Mozart K.448 on interictal epileptiform discharges in adult patients with drug-resistant focal epilepsy: A crossover randomized controlled trial

Purpose: In previous studics regarding adult patients with partial epilepsies, we statistically found that the relativc power of 8 band with power spectral analysis of their hackground EEGs showed significant inverse correlation with absolute rCBF values with 123Tc‐HMPAO SPECT or the binding potential of benzodiaapinc receptors with 1211 ‐ iomazcnil SPECT, i n each cerebral region including epileptic foci or even in the whole cerebral regions. As ii series of study aiming at iuvcstigating the utility of SPECT scans for the clinic of epilepsy, we examined with visual inspection whether thcrc were regional relationships between EEG topographies of H hand i n the background EEGs and focal ahnormalitics on 123Tc‐HMPAO SPECT or 2I‐iomazenil SPECT i n adult patients with partial epilepsies. Methods: Subjects were 27 adult epilcptic patients, composed of I4 mcn and I women, age range of 20–57 years old. Or the 27 patients, 6, 14, 2, and I patients had frontal, temporal, parietal, and occipital lobe epilepsies, respectively. The remaining 4 patients had u n tial epilepsies. Furthermore, 16 o l the 27 patients had intractable epilcpsics with seizure interval of days or weeks. All the subjects received 123Tc‐HMPAO SPECT and 99mI‐iomazenil SPECT scans and EEG examinations in the interictal periods, once an each within two weeks. MRI examinations were performed on 24 patients. EEG topographies of 0 band in background EEG activities were produced in the torm of 2‐dimensional brain mapping based on a quantitative analysis of the discontinuous monopolar recording of 24 sessions per EEG, each of which consisted of a continuous record of 5 second duration without artifacts or seizurc discharges. Results: Epileptic foci on EEG were definitely dctcrmiiicd in 24 of the 27 patients, and 18 of the 24 patients (75.0%) showed a regional agrccment between the epileptic EEG foci and the highest or higher power zones in the topographies of H band in background EEG. Thirteen of the 27 patients showed focal abnormalities consisting of hypoperfusion images in 123Tc‐HMPAO SPECT scans. Eight of the 13 patients (6 I .5%) had a regional agreement between the hypoperfusion areas and the highest or higher 0 power m i e s in EEG topographies, whereas 9 of the 13 patients (69.2%) showed a regional agreement between the hypoperfusion areas and epilcptic EEG foci. 99m1‐ iomazenil SPECT scans revealed focal hypoaccumulation images i n 17 of thc 27 patients. Twelve of thc 17 patients (70.6%) showed a regional agreement between the hypoaccumulation areas and the highest or higher H power xones i n EEG topographies, although 10 of the 17 patients (58.8%1) had a regional agreement between the hypoaccumulation areas and epileptic EEG foci. In addition, MRI showed focal organic lesions in 9 of the 24 patients, and 4 of the 9 patients (44.4%J) showed a regional agreement between the lesions and the highest or higher 0 power zones in EEG topographies.

This study was conducted to investigate the effect of low-frequency electric cortical stimulation on epileptic focus in humans. We stimulated the epileptic focus in a patient with medically intractable mesial temporal lobe epilepsy (MTLE) by means of subdural electrodes and evaluated the change in the number of interictal epileptiform discharges. We used biphasic electric current of 0.3-ms duration presented at 0.9-Hz frequency for 250 s, comparing stimulus intensity of 7.5, 2, and 0.5 mA. Interictal epileptiform discharges at the ictal focus occurred less frequently after the stimulation with the intensity of 0.5 mA. With the intensity of 7.5 mA and 2.0 mA, however, habitual auras were elicited by the stimulation, and afterdischarges were seen on the cortical EEG. Low-frequency, low-intensity electric cortical stimulation could produce inhibitory effects on epileptic activity. At the same time, however, a caution for possible induction of EEG seizures is needed, even when applying low-frequency electric stimulation.

Analysis of EEG–fMRI data in focal epilepsy based on automated spike classification and Signal Space Projection

Dynamic changes of focal hypometabolism in relation to epileptic activity

Classification of epileptic EEG signals using PSO based artificial neural network and tunable-Q wavelet transform

BackgroundThe main challenge in assessing patients with epilepsy for resective surgery is localising seizure onset. Frequently, identification of the irritative and seizure onset zones requires invasive EEG. EEG correlated functional...

Mutually suppressive interrelations of symmetric epileptic foci in bitemporal epilepsy and their inhibitory stimulation

Mutual information analysis and detection of interictal morphological differences in interictal epileptiform discharges of patients with partial epilepsies

Interictal epileptiform discharges (IEDs) are an electrographic biomarker of epilepsy. Despite their crucial role in diagnosing epilepsy, heterogeneous findings exist on the mechanisms underlying their occurrence and the effects of anti-seizure medications (ASMs) on IEDs. We conducted a study to investigate the association between IED frequency and the serum concentration of two commonly used ASMs, levetiracetam (LEV) and lamotrigine (LTG). We included 56 patients undergoing a continuous video EEG monitoring in our center with tapering of ASM. IED frequency was analyzed using automated and semiautomated methods and serum samples were collected sequentially throughout the stay. The cohort consisted of 41 patients (23 female, 18 male), between 19 and 64 years (mean 37.42 years), most of which were diagnosed with focal epilepsy (93%). IED frequency increased after ASM reduction revealing a negative correlation similarly with LEV and LTG serum concentrations (p = 0.0057 and p = 0.0426, respectively). Notably, we observed a significant increase in IED frequency following dose reduction or discontinuation of both medications. This effect was reversed after ASM were re-dosed. This may indicate the suppressive properties of LEV and LTG against epileptic seizures. Furthermore, our study highlights the importance of ASM discontinuation, which may be required for capturing IEDs during diagnostic continuous EEG monitoring, and not be fully explained by circadian or ultradian rhythms alone. Our findings contribute to the understanding of ASM effects on IED frequency dynamics and suggest seizure suppressive properties of LEV and LTG.

A central goal of the presurgical evaluation for refractory focal epilepsy is the accurate delineation of the epileptogenic zone (EZ)-the cortical region indispensable for generating epileptic seizures. Given that electroencephalography (EEG) provides superior temporal resolution while functional magnetic resonance imaging (fMRI) offers enhanced spatial localization, integrating these modalities holds significant promise for improving epileptic focus identification. In this study, we first derived characteristic spike patterns by detecting and averaging interictal epileptiform discharges (IEDs) from extraoperative EEG recordings. These patterns were then correlated with intracranial EEG data to develop an automated system for precise temporal mapping of seizure activity. Finally, we convolved the resulting temporal regressor with the hemodynamic response function (HRF) within a general linear model (GLM) framework to achieve robust localization of epileptic foci. This study was performed on five medication-resistant epilepsy patients whose neuroimaging and electrophysiological data were acquired at the National Brain Mapping Lab (NBML). Our proposed methodology demonstrated strong concordance with clinical EEG findings across all five cases. Notably, for the three surgical candidates, the approach provided additional localizing information beyond conventional EEG data. Quantitative analysis revealed statistically significant enhancements in both localization accuracy (<tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{p}&lt;0.05$</tex>) and spatial precision (mean ± SEM: <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$2.3 \pm 0.4 \text{mm}$</tex>) compared to current gold-standard techniques reported in recent literature.

One of the sectors for which Artificial Intelligence applications have been considered as exceptionally promising is the healthcare sector. As a public-facing sector, the introduction of AI applications has been subject to extended news coverage. This article conducts a quantitative and qualitative data analysis of English news media articles covering AI systems that allow the automation of tasks that so far needed to be done by a medical expert such as a doctor or a nurse thereby redistributing their agency. We investigated in this article one particular framing of AI systems and their agency: the framing that positions AI systems as (1a) replacing and (1b) outperforming the human medical expert, and in which (2) AI systems are personified and/or addressed as a person. The analysis of our data set consisting of 365 articles written between the years 1980 and 2019 will show that there is a tendency to present AI systems as outperforming human expertise. These findings are important given the central role of news coverage in explaining AI and given the fact that the popular frame of ‘outperforming’ might place AI systems above critique and concern including the Hippocratic oath. Our data also showed that the addressing of an AI system as a person is a trend that has been advanced only recently and is a new development in the public discourse about AI.

Lived experiences fundamentally shape how individuals interact with AI systems, influencing perceptions of safety, trust, and usability. While prior research has focused on developing techniques to emulate human preferences, and proposed taxonomies to categorize risks (such as psychological harms and algorithmic biases), these efforts have provided limited systematic understanding of lived human experiences or actionable strategies for embedding them meaningfully into the AI development life-cycle. This work proposes a framework for meaningfully integrating lived experience into the design and evaluation of AI systems. We synthesize interdisciplinary literature across lived experience philosophy, human-centered design, and human-AI interaction, arguing that centering lived experience can lead to models that more accurately reflect the retrospective, emotional, and contextual dimensions of human cognition. Drawing from a wide body of work across psychology, education, healthcare, and social policy, we present a targeted taxonomy of lived experiences with specific applicability to AI systems. To ground our framework, we examine three application domains— (i) education, (ii) healthcare, and (iii) cultural alignment—illustrating how lived experience informs user goals, system expectations, and ethical considerations in each context. We further incorporate insights from AI system operators and human-AI partnerships to highlight challenges in responsibility allocation, mental model calibration, and long-term system adaptation. We conclude with actionable recommendations for developing experience-centered AI systems that are not only technically robust but also empathetic, context-aware, and aligned with human realities. This work offers a foundation for future research that bridges technical development with the lived experiences of those impacted by AI systems.

PurposeInscrutable machine learning (ML) models are part of increasingly many information systems. Understanding how these models behave, and what their output is based on, is a challenge for developers let alone non-technical end users.Design/methodology/approachThe authors investigate how AI systems and their decisions ought to be explained for end users through a systematic literature review.FindingsThe authors’ synthesis of the literature suggests that AI system communication for end users has five high-level goals: (1) understandability, (2) trustworthiness, (3) transparency, (4) controllability and (5) fairness. The authors identified several design recommendations, such as offering personalized and on-demand explanations and focusing on the explainability of key functionalities instead of aiming to explain the whole system. There exists multiple trade-offs in AI system explanations, and there is no single best solution that fits all cases.Research limitations/implicationsBased on the synthesis, the authors provide a design framework for explaining AI systems to end users. The study contributes to the work on AI governance by suggesting guidelines on how to make AI systems more understandable, fair, trustworthy, controllable and transparent.Originality/valueThis literature review brings together the literature on AI system communication and explainable AI (XAI) for end users. Building on previous academic literature on the topic, it provides synthesized insights, design recommendations and future research agenda.