Face swapping in seizure videos for patient deidentification.

OBJECTIVE: This study aimed to test different AI-based face-swapping models applied to videos of epileptic seizures, with the goal of protecting patient privacy while retaining clinically useful seizure semiology. We hypothesized that specific models would show differences in semiologic fidelity compared to the original clinical videos. METHODS: Three open-source models, SimSwap, MobileFaceSwap and GHOST were adopted for face-swapping. For every model, an AI generated male and female image were used to replace the original faces. One representative seizure per patient from three patients with epilepsy was chosen (3 seizure videos x 3 AI models x 2 M/F swap) and remade to 18 transformed video clips. To evaluate the performance of the three models, we used both objective (AI-based) and subjective (expert clinician) evaluation. The objective assessment included four metrics for facial appearance and four metrics for facial expression changes. Four experienced epileptologists reviewed the clips and scoring according to deidentification and preservation of semiology. Kruskal-Wallis H test was used for statistical analysis among the models. RESULTS: In the reproduced videos, the swapped face cannot be recognized as the original face, with no significant difference in scores of deidentification either by objective or subjective assessment. Regarding semiology preservation, no significant differences between models were observed in the objective evaluations. The subjective evaluations revealed that the GHOST model outperformed the other two models (p=0.028). CONCLUSION: This is the first study evaluating AI face swapping models in epileptic seizure video clips. Optimization of AI face-swapping models could enhance the accessibility of seizure videos for education and research while protecting patient privacy and maintaining semiology.

Accepting multiple conditions in psychiatry: From comorbidity to multimorbidity.

The concept of multimorbidity in psychiatry refers to the coexistence of multiple health conditions without conceptualizing a central disorder referring to a patient-centered approach that views every diagnosis equally. It emphasizes a shift from focusing on an index disorder to considering interrelated symptoms crucial in psychiatry due to frequent multimorbidity patterns. In clinical practice, multimorbidity guides patient-centered care helping to address the holistic needs of patients and challenging the organization of mental health care. In research, multimorbidity reshapes study outcomes and promotes the development of patient-reported outcome questionnaires, also posing ethical challenges and advocating for the inclusion of multimorbid patients in personalized interventions. Finally, multimorbidity in psychiatry has public health implications aligning with the socioecological systems perspective. This notion aims to improve prognosis, care, and health costs while impacting global mortality and challenging the conceptualization and organization of psychiatric care.

Prognostic value of the 5-SENSE Score to predict focality of the seizure-onset zone as assessed by stereoelectroencephalography: a prospective international multicentre validation study.

Introduction: Epilepsy surgery is the only curative treatment for patients with drug-resistant focal epilepsy. Stereoelectroencephalography (SEEG) is the gold standard to delineate the seizure-onset zone (SOZ). However, up to 40% of patients are subsequently not operated as no focal non-eloquent SOZ can be identified. The 5-SENSE Score is a 5-point score to predict whether a focal SOZ is likely to be identified by SEEG. This study aims to validate the 5-SENSE Score, improve score performance by incorporating auxiliary diagnostic methods and evaluate its concordance with expert decisions. Methods and analysis: Non-interventional, observational, multicentre, prospective study including 200 patients with drug-resistant epilepsy aged ≥15 years undergoing SEEG for identification of a focal SOZ and 200 controls at 22 epilepsy surgery centres worldwide. The primary objective is to assess the diagnostic accuracy and generalisability of the 5-SENSE in predicting focality in SEEG in a prospective cohort. Secondary objectives are to optimise score performance by incorporating auxiliary diagnostic methods and to analyse concordance of the 5-SENSE Score with the expert decisions made in the multidisciplinary team discussion. Ethics and dissemination: Prospective multicentre validation of the 5-SENSE score may lead to its implementation into clinical practice to assist clinicians in the difficult decision of whether to proceed with implantation. This study will be conducted in accordance with the Tri-Council Policy Statement: Ethical Conduct for Research Involving Humans (2014). We plan to publish the study results in a peer-reviewed full-length original article and present its findings at scientific conferences. Trial registration number: NCT06138808.

Deep learning approaches for seizure video analysis: A review.

Seizure events can manifest as transient disruptions in the control of movements which may be organized in distinct behavioral sequences, accompanied or not by other observable features such as altered facial expressions. The analysis of these clinical signs, referred to as semiology, is subject to observer variations when specialists evaluate video-recorded events in the clinical setting. To enhance the accuracy and consistency of evaluations, computer-aided video analysis of seizures has emerged as a natural avenue. In the field of medical applications, deep learning and computer vision approaches have driven substantial advancements. Historically, these approaches have been used for disease detection, classification, and prediction using diagnostic data; however, there has been limited exploration of their application in evaluating video-based motion detection in the clinical epileptology setting. While vision-based technologies do not aim to replace clinical expertise, they can significantly contribute to medical decision-making and patient care by providing quantitative evidence and decision support. Behavior monitoring tools offer several advantages such as providing objective information, detecting challenging-to-observe events, reducing documentation efforts, and extending assessment capabilities to areas with limited expertise. The main applications of these could be (1) improved seizure detection methods; (2) refined semiology analysis for predicting seizure type and cerebral localization. In this paper, we detail the foundation technologies used in vision-based systems in the analysis of seizure videos, highlighting their success in semiology detection and analysis, focusing on work published in the last 7 years. We systematically present these methods and indicate how the adoption of deep learning for the analysis of video recordings of seizures could be approached. Additionally, we illustrate how existing technologies can be interconnected through an integrated system for video-based semiology analysis. Each module can be customized and improved by adapting more accurate and robust deep learning approaches as these evolve. Finally, we discuss challenges and research directions for future studies.

Generalisability of epileptiform patterns across time and patients.

The complexity of localising the epileptogenic zone (EZ) contributes to surgical resection failures in achieving seizure freedom. The distinct patterns of epileptiform activity during interictal and ictal phases, varying across patients, often lead to suboptimal localisation using electroencephalography (EEG) features. We posed two key questions: whether neural signals reflecting epileptogenicity generalise from interictal to ictal time windows within each patient, and whether epileptiform patterns generalise across patients. Utilising an intracranial EEG dataset from 55 patients, we extracted a large battery of simple to complex features from stereo-EEG (SEEG) and electrocorticographic (ECoG) neural signals during interictal and ictal windows. Our features (n = 34) quantified many aspects of the signals including statistical moments, complexities, frequency-domain and cross-channel network attributes. Decision tree classifiers were then trained and tested on distinct time windows and patients to evaluate the generalisability of epileptogenic patterns across time and patients, respectively. Evidence strongly supported generalisability from interictal to ictal time windows across patients, particularly in signal power and high-frequency network-based features. Consistent patterns of epileptogenicity were observed across time windows within most patients, and signal features of epileptogenic regions generalised across patients, with higher generalisability in the ictal window. Signal complexity features were particularly contributory in cross-patient generalisation across patients. These findings offer insights into generalisable features of epileptic neural activity across time and patients, with implications for future automated approaches to supplement other EZ localisation methods.

Grading system for assessing the confidence in the epileptogenic zone reported in published studies: A Delphi consensus study.

OBJECTIVE: This study was undertaken to develop a standardized grading system based on expert consensus for evaluating the level of confidence in the localization of the epileptogenic zone (EZ) as reported in published studies, to harmonize and facilitate systematic reviews in the field of epilepsy surgery. METHODS: We conducted a Delphi study involving 22 experts from 18 countries, who were asked to rate their level of confidence in the localization of the EZ for various theoretical clinical scenarios, using different scales. Information provided in these scenarios included one or several of the following data: magnetic resonance imaging (MRI) findings, invasive electroencephalography summary, and postoperative seizure outcome. RESULTS: The first explorative phase showed an overall interrater agreement of .347, pointing to large heterogeneity among experts' assessments, with only 17% of the 42 proposed scenarios associated with a substantial level of agreement. A majority showed preferences for the simpler scale and single-item scenarios. The successive Delphi voting phases resulted in a majority consensus across experts, with more than two thirds of respondents agreeing on the rating of each of the tested single-item scenarios. High or very high levels of confidence were ascribed to patients with either an Engel class I or class IA postoperative seizure outcome, a well-delineated EZ according to all available invasive EEG (iEEG) data, or a well-delineated focal epileptogenic lesion on MRI. MRI signs of hippocampal sclerosis or atrophy were associated with a moderate level of confidence, whereas a low level was ascribed to other MRI findings, a poorly delineated EZ according to iEEG data, or an Engel class II-IV postoperative seizure outcome. SIGNIFICANCE: The proposed grading system, based on an expert consensus, provides a simple framework to rate the level of confidence in the EZ reported in published studies in a structured and harmonized way, offering an opportunity to facilitate and increase the quality of systematic reviews and guidelines in the field of epilepsy surgery.

Stress and Epilepsy: Towards Understanding of Neurobiological Mechanisms for Better Management.

Stress has been identified as a major contributor to human disease and is postulated to play a substantial role in epileptogenesis. In a significant proportion of individuals with epilepsy, sensitivity to stressful events contributes to dynamic symptomatic burden, notably seizure occurrence and frequency, and presence and severity of psychiatric comorbidities [anxiety, depression, posttraumatic stress disorder (PTSD)]. Here, we review this complex relationship between stress and epilepsy using clinical data and highlight key neurobiological mechanisms including the hypothalamic-pituitary-adrenal (HPA) axis dysfunction, altered neuroplasticity within limbic system structures, and alterations in neurochemical pathways such as brain-derived neurotrophic factor (BNDF) linking epilepsy and stress. We discuss current clinical management approaches of stress that help optimize seizure control and prevention, as well as psychiatric comorbidities associated with epilepsy. We propose that various shared mechanisms of stress and epilepsy present multiple avenues for the development of new symptomatic and preventative treatments, including disease modifying therapies aimed at reducing epileptogenesis. This would require close collaborations between clinicians and basic scientists to integrate data across multiple scales, from genetics to systems biology, from clinical observations to fundamental mechanistic insights. In future, advances in machine learning approaches and neuromodulation strategies will enable personalized and targeted interventions to manage and ultimately treat stress-related epileptogenesis.

BDNF as potential biomarker of epilepsy severity and psychiatric comorbidity: pitfalls in the clinical population.

BACKGROUND: Several studies implicate brain-derived neurotrophic factor (BDNF) in the pathophysiology of epilepsy. In particular, preclinical data suggest that lower serum BDNF is a biomarker of epilepsy severity and psychiatric comorbidities. We tested this prediction in clinical epilepsy cohorts. METHODS: Patients with epilepsy were recruited from 4 epilepsy centers in France and serum BDNF was quantified. Clinical characteristics including epilepsy duration, classification, localization, etiology, seizure frequency and drug resistance were documented. Presence of individual anti-seizure medications (ASM) was noted. Screening for depression and anxiety symptoms was carried out in all patients using the NDDI-E and the GAD-7 scales. In patients with positive screening for anxiety and/or depression, detailed psychiatric testing was performed including the Mini International Neuropsychiatric Interview (MINI), STAI-Y, Holmes Rahe Stressful Events Scale and Beck Depression Interview. Descriptive analysis was applied. Spearman's test and Pearson's co-efficient were used to assess the association between BDNF level and continuous variables. For discrete variables, comparison of means (Student's t-test, Mann-Whitney u-test) was used to compare mean BDNF serum level between groups. Multivariate analysis was performed using a regression model. RESULTS: No significant correlation was found between serum BDNF level and clinical features of epilepsy or measures of depression. The main group-level finding was that presence of any ASM at was associated with increased BDNF; this effect was particularly significant for valproate and perampanel. CONCLUSION: Presence of ASM affects serum BDNF levels in patients with epilepsy. Future studies exploring BDNF as a possible biomarker of epilepsy severity and/or psychiatric comorbidity must control for ASM effects.

Post-traumatic stress disorder (PTSD), cognitive control, and perceived seizure control in patients with epilepsy: An exploratory study.

Epilepsy is often linked to various psychiatric symptoms, with anxiety, depression, and interictal dysphoric disorders being the most prevalent. Few studies have investigated posttraumatic stress disorder (PTSD) in epilepsy, but they suggest a notable prevalence of PTSD. PTSD is known to be associated with cognitive impairments, particularly memory and executive functions. Our proposed exploratory study aims to investigate executive attentional control and emotional inhibition in patients with drug-resistant epilepsy (DRE) who exhibit PTSD symptoms compared with a healthy control group. Additionally, some PWE can manage their seizures using emotional and cognitive strategies, we find it relevant to explore the connection between their regulation abilities, cognitive control performance, and PTSD symptoms. We included 54 PWE and 60 healthy participants. They completed anxiety and depression scales as well as two questionnaires assessing PTSD symptoms and a questionnaire that measured the perceived self-control of seizures. We measured executive control using an executive control task (Attention Network Test, ANT) and an emotional Go/No-Go task. We found a positive correlation between PTSD scores (PDS-5) and performance at the ANT task. In contrast, in the emotional inhibition (Go/No-Go) task, behavioral inhibition errors were positively correlated with PTSD scores, specifically with hypervigilance symptoms in PTSD+ patients. There was a positive correlation between response reaction times in an aversive condition and PTSD scores: the more severe the PTSD symptoms, the faster the PWE identified stimuli in the angry face condition of the Go/No-Go task. Regarding perceived seizure control, we found correlations between alertness and PTSD symptoms associated with seizure anticipation during the inter- and peri-ictal periods. Patients with PTSD symptoms reported better seizure control. Our findings suggest that epilepsy patients with PTSD experience cognitive changes such as heightened executive attentional control, weakened emotional inhibition, and improved seizure control perception.

"Plok-plok" syndrome: Posttraumatic stress disorder following an SEEG thermocoagulation and direct electrical stimulation procedure.

The psychological impact of intracerebral electroencephalography (stereoelectroencephalography [SEEG]) including the thermocoagulation procedure has not yet been clearly studied. We present a case of a patient who, following an SEEG procedure for presurgical evaluation of intractable focal epilepsy, developed severe symptoms of posttraumatic stress disorder. Such an occurrence may be under-estimated. Perceived traumatic exposure during SEEG and the development of posttraumatic psychological symptoms should be further studied in order to define risk factors and to improve the monitoring and psychological management of patients during their hospitalization. A careful and systematic procedure of prevention and support before, during, and after SEEG could decrease the risk of development or worsening of symptoms of anxiety, depression, and posttraumatic stress disorder.

Radiosurgical Corpus Callosotomy for Intractable Epilepsy: Retrospective Long-Term Safety and Efficacy Assessment in 19 Patients an Review of the Literature.

BACKGROUND: Some patients suffering from intractable epileptic seizures, particularly drop attacks (DAs), are nonremediable by curative techniques. Palliative procedure carries a significant rate of surgical and neurological complications. OBJECTIVE: To propose evaluation of safety and efficacy of Gamma Knife corpus callosotomy (GK-CC) as an alternative to microsurgical corpus callosotomy. METHODS: This study included retrospective analysis of 19 patients who underwent GK-CC between 2005 and 2017. RESULTS: Of the 19 patients, 13 (68%) had improvement in seizure control and 6 had no significant improvement. Of the 13/19 (68%) with improvement in seizures, 3 (16%) became completely seizure-free, 2 (11%) became free of DA and generalized tonic-clonic but with residual other seizures, 3 (16%) became free of DA only, and 5 (26%) had >50% reduction in frequency of all seizure types. In the 6 (31%) patients with no appreciable improvement, there were residual untreated commissural fibers and incomplete callosotomy rather than failure of Gamma Knife to disconnect. Seven patients showed a transient mild complication (37% of patients, 33% of the procedures). No permanent complication or neurological consequence was observed during the clinical and radiological workup with a mean of 89 (42-181) months, except 1 patient who had no improvement of epilepsy and then aggravation of the pre-existing cognitive and walking difficulties (Lennox-Gastaut). The median time of improvement after GK-CC was 3 (1-6) months. CONCLUSION: Gamma Knife callosotomy is safe and accurate with comparable efficacy to open callosotomy in this cohort of patients with intractable epilepsy suffering from severe drop attacks.

Potential new roles for glycogen in epilepsy.

Seizures often originate in epileptogenic foci. Between seizures (interictally), these foci and some of the surrounding tissue often show low signals with 18 fluorodeoxyglucose (FDG) positron emission tomography (PET) in many epileptic patients, even when there are no radiologically detectable structural abnormalities. Low FDG-PET signals are thought to reflect glucose hypometabolism. Here, we review knowledge about metabolism of glucose and glycogen and oxidative stress in people with epilepsy and in acute and chronic rodent seizure models. Interictal brain glucose levels are normal and do not cause apparent glucose hypometabolism, which remains unexplained. During seizures, high amounts of fuel are needed to satisfy increased energy demands. Astrocytes consume glycogen as an additional emergency fuel to supplement glucose during high metabolic demand, such as during brain stimulation, stress, and seizures. In rodents, brain glycogen levels drop during induced seizures and increase to higher levels thereafter. Interictally, in people with epilepsy and in chronic epilepsy models, normal glucose but high glycogen levels have been found in the presumed brain areas involved in seizure generation. We present our new hypothesis that as an adaptive response to repeated episodes of high metabolic demand, high interictal glycogen levels in epileptogenic brain areas are used to support energy metabolism and potentially interictal neuronal activity. Glycogenolysis, which can be triggered by stress or oxidative stress, leads to decreased utilization of plasma glucose in epileptogenic brain areas, resulting in low FDG signals that are related to functional changes underlying seizure onset and propagation. This is (partially) reversible after successful surgery. Last, we propose that potential interictal glycogen depletion in epileptogenic and surrounding areas may cause energy shortages in astrocytes, which may impair potassium buffering and contribute to seizure generation. Based on these hypotheses, auxiliary fuels or treatments that support glycogen metabolism may be useful to treat epilepsy.

Interictal HFO and FDG-PET correlation predicts surgical outcome following SEEG.

OBJECTIVE: This study aimed to investigate the quantitative relationship between interictal 18 F-fluorodeoxyglucose-positron emission tomography (FDG-PET) and interictal high-frequency oscillations (HFOs) from stereo-electroencephalography (SEEG) recordings in patients with refractory epilepsy. METHODS: We retrospectively included 32 patients. FDG-PET data were quantified through statistical parametric mapping (SPM) t test modeling with normal controls. Interictal SEEG segments with four, 10-min segments were selected randomly. HFO detection and classification procedures were automatically performed. Channel-based HFOs separating ripple (80-250 Hz) and fast ripple (FR; 250-500 Hz) counts were correlated with the surrounding metabolism T score at the individual and group level, respectively. The association was further validated across anatomic seizure origins and sleep vs wake states. We built a joint feature FR × T reflecting the FR and hypometabolism concordance to predict surgical outcomes in 28 patients who underwent surgery. RESULTS: We found a negative correlation between interictal FDG-PET and HFOs through the linear mixed-effects model (R2  = .346 and .457 for ripples and FRs, respectively, p < .001); these correlations were generalizable to different epileptogenic-zone lobar localizations and vigilance states. The FR × T inside the resection volume could be used as a predictor for surgical outcomes with an area under the curve of 0.81. SIGNIFICANCE: The degree of hypometabolism is associated with HFO generation rate, especially for FRs. This relationship would be meaningful for selection of SEEG candidates and for optimizing SEEG scheme planning. The concordance between FRs and hypometabolism inside the resection volume could provide prognostic information regarding surgical outcome.

Characteristics and Neural Correlates of Emotional Behavior during Prefrontal Seizures.

OBJECTIVE: This study was undertaken to characterize clinical expression and intracerebral electroencephalographic (EEG) correlates of emotional expression during prefrontal epileptic seizures. METHODS: We performed a descriptive analysis of seizure semiology in patients explored with stereo-EEG (SEEG) for pharmacoresistant prefrontal epilepsy, using a semiquantitative score for seizure-related emotional behavior. Two independent observers scored occurrence and intensity of objective emotional features (face/body movements/vocalization/overall appearance), testing interobserver reliability. Intracerebral electrophysiological changes were documented. Cluster analysis and principal component analysis (PCA) compared behavioral signs with neural SEEG correlates. For each patient, the clinical and anatomoelectrophysiological scores were established, based on a prototypical emotional seizure. RESULTS: Forty-two patients (469 seizures) were included. Interobserver correlation for emotional signs was satisfactory (kappa = 0.6-0.8). Prevalence of any subjective and/or objective ictal emotional phenomena was 79% (33/42); objective emotional signs occurred in 27 of 42 subjects (64%). Negatively valenced emotional semiology (ictal feeling of fear, defensive and/or aggressive behaviors) was much more prevalent than positively valenced, prosocial behaviors. Cluster analysis and PCA identified 4 groups with different occurrence of emotional signs and cerebral correlates. Two main clusters of negatively valenced behavior were identified: "active threat response," associated with seizure organizations involving posterior orbitofrontal cortex, anterior cingulate, and dorsolateral and/or ventrolateral prefrontal cortex; and "passive fear," associated with amygdala, other mesial temporal structures, and posterior orbitofrontal cortex. INTERPRETATION: Emotional behaviors, especially fear/threat response, are common in prefrontal seizures, reflecting the role of the prefrontal cortex in emotional control. Different cortical seizure localizations were associated with "passive fear" and "active threat response" seizure behaviors at the group level. ANN NEUROL 2022;92:1052-1065.

Automated video analysis of emotion and dystonia in epileptic seizures.

OBJECTIVE: To investigate the accuracy of deep learning methods applied to seizure video data, in discriminating individual semiologic features of dystonia and emotion in epileptic seizures. METHODS: A dataset of epileptic seizure videos was used from patients explored with stereo-EEG for focal pharmacoresistant epilepsy. All patients had hyperkinetic (HKN) seizures according to ILAE definition. Presence or absence of (1) dystonia and (2) emotional features in each seizure was documented by an experienced clinician. A deep learning multi-stream model with appearance and skeletal keypoints, face and body information, using graph convolutional neural networks, was used to test discrimination of dystonia and emotion. Classification accuracy was assessed using a leave-one-subject-out analysis. RESULTS: We studied 38 HKN seizure videos in 19 patients. By visual analysis based on ILAE criteria, 9/19 patients were considered to have dystonia and 9/19 patients were considered to have emotional signs. Two patients had both dystonia and emotional signs. Applying the deep learning multistream model, spatiotemporal features of facial appearance showed best accuracy for emotion detection (F1 score 0.84), while skeletal keypoint detection performed best for dystonia (F1 score 0.83). SIGNIFICANCE: Here, we investigated deep learning of video data for analyzing individual semiologic features of dystonia and emotion in hyperkinetic seizures. Automated classification of individual semiologic features is possible and merits further study.