Background
Psychogenic non‐epileptic seizure‐status (PNES‐status), defined by psychogenic non‐epileptic seizures (PNES) over 30 min, are often misdiagnosed as status epilepticus. We aimed to describe ...the features of patients who experienced PNES‐status, admitted to an intensive care unit (ICU).
Methods
We screened the patients hospitalized in our epilepsy unit during a 4‐year period, with a diagnosis of PNES‐status and ICU admission.
Results
Among 171 patients with PNES, we identified 25 patients (15%) who presented 39 episodes of PNES‐status leading to ICU admission. Some 76% of the patients were women. The median age at the time of the PNES‐status episode was 35 years. Half (48%) alleged a history of epilepsy, but epilepsy was confirmed in only 12%. A history of psychiatric disease was found in 68%. PNES were present in 85% of patients before PNES‐status, and semiology of PNES and PNES‐status was similar for 79% of the patients, including hyperkinetic movements in 95% of the episodes and suspected loss of consciousness in 87%. Benzodiazepines were administrated in 77% of the episodes, antiepileptic drugs in 87%, and antibiotherapy for a ICU‐related infection in 15% of the episodes. Oral intubation was performed in 41% of the episodes. Blood tests showed normal levels of creatine phosphokinase and leucocytes in 90% and 95% of the episodes, respectively. No epileptic activity was found during per‐event electroencephalography but interictal epileptic activity was found in 10% of the episodes.
Conclusion
Hyperkinetic PNES‐status should always be considered as a differential diagnosis of status epilepticus, with a high risk of iatrogenic consequences.
Psychogenic non epileptic seizure‐status (PNES‐Status) can be misdiagnosed as status epilepticus. Among 171 patients with PNES, we identified 25 patients who experienced 39 episodes of PNES‐status leading to intensive care unit admission. 48% alleged a history of epilepsy but epilepsy was confirmed in only 12%. The clinical presentation was hyperkinetic movements and a suspected loss of consciousness. Benzodiazepines were administrated in 77% of the episodes, and antiepileptic drugs in 87%. Oral intubation was performed in 41%. PNES‐status should always be considered as differential diagnosis of status epilepticus, because of a high risk of iatrogenic consequences.
•Two major improvements on a noise-assisted EMD method are proposed.•The obtained components contain less noise and more physical meaning.•These improvements are confirmed on artificial signals.•The ...new method is successfully tested on several biomedical signals.•Information of physiological phenomena involved in their production is obtained.
The empirical mode decomposition (EMD) decomposes non-stationary signals that may stem from nonlinear systems, in a local and fully data-driven manner. Noise-assisted versions have been proposed to alleviate the so-called “mode mixing” phenomenon, which may appear when real signals are analyzed. Among them, the complete ensemble EMD with adaptive noise (CEEMDAN) recovered the completeness property of EMD. In this work we present improvements on this last technique, obtaining components with less noise and more physical meaning. Artificial signals are analyzed to illustrate the capabilities of the new method. Finally, several real biomedical signals are decomposed, obtaining components that represent physiological phenomenons.
•CNN and LSTM are integrated for efficient detection of the epileptic seizure.•A comparison among time, frequency and time–frequency features are presented.•A highest classification accuracy of ...99.27% is achieved.•The proposed method is useful for binary and multi-class classification problems.
Advances in deep learning methods present new opportunities for fixing complex problems for an end to end learning. In terms of optimal design, seizure detection from EEG data has not been completely exploited by current models of deep learning. Most of the previous studies focus on handcrafted feature extraction for seizure detection. However, this method is not generalizable and needs major changes inside a new dataset for each new patient. In this paper, we proposed autonomously generalized retrospective and patient-specific hybrid models based on two types of feature extractors, namely Convolutional Neural Networks along with long short-term memory. The model automatically generates customized features to better classify ictal, interictal, and preictal segments for each patient and make it ideal for real-time. The procedure can be extended to any patient from Freiburg epileptic seizure database without the need for manual feature extraction. The method decomposed the EEG signals into time-based, frequency-based, and time–frequency-based features that were tested and compared in 21 subjects. Three forms of experiments including two binary classification problems and a ternary classification were performed to investigate the feasibility of the proposed approach. Using the time–frequency domain signals an average accuracy of 99.19%, 99.27%, and 95.04%, with frequency-domain signals, average accuracies of 96.64%, 95.75%, and 93.42% while with time-domain signals an average accuracy of 94.71%, 93.99%, and 90.53% was obtained. Our work shows that the combined use of CNNs and LSTMs by integrating spatial and temporal context along with time–frequency domain signals can significantly improve the accuracy of seizure detection.
Na
v
1.1 is a major Na
v
subtype in the brain. Up to 900 nonsense and missense mutations in
SCN1A
, the coding gene for Na
v
1.1, have been identified in patients with epilepsy syndromes. Here, we ...report the cryo-EM structure of the human Na
v
1.1–β4 complex. Comparative structural analysis of hundreds of missense disease mutations in Na
v
1.1 and Na
v
1.5, whose structure is reported in the accompanying paper, reveals 70 loci that are common in these two channels. Several clusters, defined as the mutational hotspots, are identified and generally classified as the structural mutations and functional mutations. Our comparative structural analyses establish a framework for structure–function relationship dissection of Na
v
disease variants and will facilitate development of precision medicine for various sodium channelopathies.
Among the nine subtypes of human voltage-gated sodium (Na
v
) channels, the brain and cardiac isoforms, Na
v
1.1 and Na
v
1.5, each carry more than 400 missense mutations respectively associated with epilepsy and cardiac disorders. High-resolution structures are required for structure–function relationship dissection of the disease variants. We report the cryo-EM structures of the full-length human Na
v
1.1–β4 complex at 3.3 Å resolution here and the Na
v
1.5-E1784K variant in the accompanying paper. Up to 341 and 261 disease-related missense mutations in Na
v
1.1 and Na
v
1.5, respectively, are resolved. Comparative structural analysis reveals several clusters of disease mutations that are common to both Na
v
1.1 and Na
v
1.5. Among these, the majority of mutations on the extracellular loops above the pore domain and the supporting segments for the selectivity filter may impair structural integrity, while those on the pore domain and the voltage-sensing domains mostly interfere with electromechanical coupling and fast inactivation. Our systematic structural delineation of these mutations provides important insight into their pathogenic mechanism, which will facilitate the development of precise therapeutic interventions against various sodium channelopathies.
Recent advancements in Deep Learning models hold the potential to revolutionize the automated analysis of EEG data for early and accurate diagnosis of epileptic seizures. This paper introduces an ...interpretable hybrid model, integrating Convolutional Neural Networks (CNNs) and Long Short-Term Memory (LSTM) networks, with a primary emphasis on two-class epileptic seizure classification. Extensive evaluations across diverse datasets establish the model's resilience and effectiveness. Notably, in the CHB-MIT dataset, the model achieves an average validation accuracy of 92.8 %, an average area under the curve of 93 %, an average specificity of 90.3 %, an average sensitivity of 95 %, an F1-score of 94 %, and an MCC of 88.2 %. In the Siena dataset, an average validation accuracy of 92.7 %, an average area under the curve of 93 %, an average specificity of 84 %, an average sensitivity of 91 %, an F1-score of 92.5 %, and an MCC of 85 % are maintained. In the Helsinki dataset, the model attains an average validation accuracy of 86.4 %, accompanied by an average area under the curve of 86 %, an average specificity of 84 %, a sensitivity of 88 %, an F1-score of 87.8 %, and an MCC of 75.3 %.
Furthermore, the proposed model provides a post-hoc explainer utilizing the Shapley Additive Explanations (SHAP) method, specifically the SHAP Gradient Explainer that interprets the predictive model by providing two forms of explanation: (i) Event-wise explanations, elucidating why particular EEG data segments are classified as seizures or non-seizure events, and (ii) Patient-wise explanations that precisely pinpoint the brain lobe and hemisphere responsible for the seizure's origin. The explainer's efficacy is meticulously assessed using ground truth data, yielding localization and lateralization accuracy scores of 85.43 % for the CHB-MIT dataset, 86 % for the Siena dataset, and 79.4 % for the Helsinki dataset.
This research contributes to the advancement of the responsible and trustworthy use of Artificial Intelligence in seizure vs. non-seizure EEG classification and interpretation, delivering both precise classification and in-depth explanations.
Electroencephalogram (EEG) signals play an important role in the epilepsy detection. In the past decades, the automatic detection system of epilepsy has emerged and performed well. In this paper, a ...novel sparse representation-based epileptic seizure classification based on the dictionary learning with homotopy (DLWH) algorithm is proposed. The performance of the proposed method evaluates on two public EEG databases provided by the Bonn University and Childrens Hospital Boston-Massachusetts Institute of Technology (CHB-MIT), various classification cases that include health and seizure; non-seizure and seizure; inter ictal (seizure-free interval) and ictal (seizure). The results show that the DLWH only completes the test with 19.671 s compared with the traditional sparse representation methods with high degree of automation, which are better than those obtained using the well-known dictionary learning method. Besides, two publicly available benchmark databases recognition rates are as high as 100%, 99.5%, with average of 99.5% and 95.06%,% respectively, and the results show that the epileptic detection system based on the dictionary learning has a high application value.
Epilepsy is a neurological disorder in which involuntary contractions, sensory abnormalities, and changes occur as a result of abrupt and uncontrolled discharges in the neurons in the brain, which ...disrupt the systems regulated by the brain. In epilepsy, abnormal electrical impulses from cells in various brain areas are noticed. The accurate interpretation of these electrical impulses is critical in the illness diagnosis. This study aims to use different machine-learning algorithms to diagnose epileptic seizures. The frequency components of EEG data were extracted using parametric approaches. This feature extraction approach was fed into machine learning classification algorithms, including Artificial Neural Network (ANN), Gradient Boosting, and Random Forest. The ANN classifier was shown to have the most significant test performance in this investigation, with roughly 97% accuracy and a 91% F1 score in recognizing epileptic episodes. The Gradient Boosting classifier, on the other hand, performed similarly to the ANN, with 96% accuracy and a 93% F1 score.
Electroencephalogram (EEG) plays a pivotal role in the detection and analysis of epileptic seizures, which affects over 70 million people in the world. Nonetheless, the visual interpretation of EEG ...signals for epilepsy detection is laborious and time-consuming. To tackle this open challenge, we introduce a straightforward yet efficient hybrid deep learning approach, named ResBiLSTM, for detecting epileptic seizures using EEG signals. Firstly, a one-dimensional residual neural network (ResNet) is tailored to adeptly extract the local spatial features of EEG signals. Subsequently, the acquired features are input into a bidirectional long short-term memory (BiLSTM) layer to model temporal dependencies. These output features are further processed through two fully connected layers to achieve the final epileptic seizure detection. The performance of ResBiLSTM is assessed on the epileptic seizure datasets provided by the University of Bonn and Temple University Hospital (TUH). The ResBiLSTM model achieves epileptic seizure detection accuracy rates of 98.88–100% in binary and ternary classifications on the Bonn dataset. Experimental outcomes for seizure recognition across seven epilepsy seizure types on the TUH seizure corpus (TUSZ) dataset indicate that the ResBiLSTM model attains a classification accuracy of 95.03% and a weighted F1 score of 95.03% with 10-fold cross-validation. These findings illustrate that ResBiLSTM outperforms several recent deep learning state-of-the-art approaches.
•PS-PTSD is an epilepsy specific psychiatric disorder.•The interaction between inter ictal and peri ictal anxiety appear to have a critical role in PS-PTSD.•History of past traumatic experiences ...seems to be linked with PS-PTSD through interictal depression and anxiety symptoms.
Previous studies investigated the varying prevalence of post-epileptic seizure posttraumatic stress disorder (PS-PTSD). The current study aimed first to compare the profiles of patients with and without PS-PTSD and, second, to study the interaction between other past traumatic experiences, subjective ictal anxiety, psychiatric comorbidities, and PS-PTSD in people with epilepsy (PWE).
We conducted an observational study, investigating past traumatic experiences and PS-PTSD through standardized scales (CTQ-28, LEC-5 and PCL-5). We used semi-structured interviews and validated psychometric scales (NDDIE for depression and GAD-7 for anxiety) to collect data on general psychiatric comorbidities. We also assessed epilepsy specific psychiatric symptoms (interictal and peri-ictal). We performed a mediation analysis through PROCESS for SPSS to evaluate the effect of history of past trauma and subjective ictal anxiety on PS-PTSD through interictal depression and anxiety symptoms.
We enrolled 135 PWE, including 35 patients with PS-PTSD (29.5 %). Patients with PS-PTSD had significantly higher depression (12.87 vs 10; p = 0.005) and anxiety (7.74 vs 5.01; p = 0.027) scores and higher prevalence of peri-ictal psychiatric symptoms, compared to patients without PS-PTSD. The relationship between other past traumatic experiences and PS-PTSD was totally mediated by interictal depression and anxiety. We found a significant indirect effect of interictal anxiety symptoms on the path between subjective ictal anxiety and PS-PTSD.
Our results showed that patients with PS-PTSD have a more severe psychopathological profile (more peri ictal and inter ictal depressive and anxiety symptoms). Both inter ictal and subjective ictal anxiety appear to have a significant role in PS-PTSD.