Diagnostic biomarker discovery from brain EEG data using LSTM, reservoir-SNN, and NeuCube methods in a pilot study comparing epilepsy and migraine.

The study introduces a new online spike encoding algorithm for spiking neural networks (SNN) and suggests new methods for learning and identifying diagnostic biomarkers using three prominent deep learning neural network models: deep BiLSTM, reservoir SNN, and NeuCube. EEG data from datasets related to epilepsy, migraine, and healthy subjects are employed. Results reveal that BiLSTM hidden neurons capture biological significance, while reservoir SNN activities and NeuCube spiking dynamics identify EEG channels as diagnostic biomarkers. BiLSTM and reservoir SNN achieve 90 and 85% classification accuracy, while NeuCube achieves 97%, all methods pinpointing potential biomarkers like T6, F7, C4, and F8. The research bears implications for refining online EEG classification, analysis, and early brain state diagnosis, enhancing AI models with interpretability and discovery. The proposed techniques hold promise for streamlined brain-computer interfaces and clinical applications, representing a significant advancement in pattern discovery across the three most popular neural network methods for addressing a crucial problem. Further research is planned to study how early can these diagnostic biomarkers predict an onset of brain states.

The value of linear and non-linear quantitative EEG analysis in paediatric epilepsy surgery: a machine learning approach.

Epilepsy surgery is effective for patients with medication-resistant seizures, however 20-40% of them are not seizure free after surgery. Aim of this study is to evaluate the role of linear and non-linear EEG features to predict post-surgical outcome. We included 123 paediatric patients who underwent epilepsy surgery at Bambino Gesù Children Hospital (January 2009-April 2020). All patients had long term video-EEG monitoring. We analysed 1-min scalp interictal EEG (wakefulness and sleep) and extracted 13 linear and non-linear EEG features (power spectral density (PSD), Hjorth, approximate entropy, permutation entropy, Lyapunov and Hurst value). We used a logistic regression (LR) as feature selection process. To quantify the correlation between EEG features and surgical outcome we used an artificial neural network (ANN) model with 18 architectures. LR revealed a significant correlation between PSD of alpha band (sleep), Mobility index (sleep) and the Hurst value (sleep and awake) with outcome. The fifty-four ANN models gave a range of accuracy (46-65%) in predicting outcome. Within the fifty-four ANN models, we found a higher accuracy (64.8% ± 7.6%) in seizure outcome prediction, using features selected by LR. The combination of PSD of alpha band, mobility and the Hurst value positively correlate with good surgical outcome.

Approach to Patients with Seizures and Epilepsy: A Guide for Primary Care Physicians.

Seizures and epilepsy are common neurologic conditions that are frequently encountered in the outpatient primary care setting. An accurate diagnosis relies on a thorough clinical history and evaluation. Understanding seizure semiology and classification is crucial in conducting the initial assessment. Knowledge of common seizure triggers and provoking factors can further guide diagnostic testing and initial management. The pharmacodynamic characteristics and side effect profiles of anti-seizure medications are important considerations when deciding treatment and counseling patients, particularly those with comorbidities and in special populations such as patient of childbearing potential.

Performance investigation of epilepsy detection from noisy EEG signals using base-2-meta stacking classifier.

Epilepsy is a chronic neurological disease, characterized by spontaneous, unprovoked, recurrent seizures that may lead to long-term disability and premature death. Despite significant efforts made to improve epilepsy detection clinically and pre-clinically, the pervasive presence of noise in EEG signals continues to pose substantial challenges to their effective application. In addition, discriminant features for epilepsy detection have not been investigated yet. The objective of this study is to develop a hybrid model for epilepsy detection from noisy and fragmented EEG signals. We hypothesized that a hybrid model could surpass existing single models in epilepsy detection. Our approach involves manual noise rejection and a novel statistical channel selection technique to detect epilepsy even from noisy EEG signals. Our proposed Base-2-Meta stacking classifier achieved notable accuracy (0.98 ± 0.05), precision (0.98 ± 0.07), recall (0.98 ± 0.05), and F1 score (0.98 ± 0.04) even with noisy 5-s segmented EEG signals. Application of our approach to the specific problem like detection of epilepsy from noisy and fragmented EEG data reveals a performance that is not only superior to others, but also is translationally relevant, highlighting its potential application in a clinic setting, where EEG signals are often noisy or scanty. Our proposed metric DF-A (Discriminant feature-accuracy), for the first time, identified the most discriminant feature with models that give A accuracy or above (A = 95 used in this study). This groundbreaking approach allows for detecting discriminant features and can be used as potential electrographic biomarkers in epilepsy detection research. Moreover, our study introduces innovative insights into the understanding of these features, epilepsy detection, and cross-validation, markedly improving epilepsy detection in ways previously unavailable.

Differentiating Epileptic and Psychogenic Non-Epileptic Seizures Using Machine Learning Analysis of EEG Plot Images.

The treatment of epilepsy, the second most common chronic neurological disorder, is often complicated by the failure of patients to respond to medication. Treatment failure with anti-seizure medications is often due to the presence of non-epileptic seizures. Distinguishing non-epileptic from epileptic seizures requires an expensive and time-consuming analysis of electroencephalograms (EEGs) recorded in an epilepsy monitoring unit. Machine learning algorithms have been used to detect seizures from EEG, typically using EEG waveform analysis. We employed an alternative approach, using a convolutional neural network (CNN) with transfer learning using MobileNetV2 to emulate the real-world visual analysis of EEG images by epileptologists. A total of 5359 EEG waveform plot images from 107 adult subjects across two epilepsy monitoring units in separate medical facilities were divided into epileptic and non-epileptic groups for training and cross-validation of the CNN. The model achieved an accuracy of 86.9% (Area Under the Curve, AUC 0.92) at the site where training data were extracted and an accuracy of 87.3% (AUC 0.94) at the other site whose data were only used for validation. This investigation demonstrates the high accuracy achievable with CNN analysis of EEG plot images and the robustness of this approach across EEG visualization software, laying the groundwork for further subclassification of seizures using similar approaches in a clinical setting.

Calibrating Deep Learning Classifiers for Patient-Independent Electroencephalogram Seizure Forecasting.

The recent scientific literature abounds in proposals of seizure forecasting methods that exploit machine learning to automatically analyze electroencephalogram (EEG) signals. Deep learning algorithms seem to achieve a particularly remarkable performance, suggesting that the implementation of clinical devices for seizure prediction might be within reach. However, most of the research evaluated the robustness of automatic forecasting methods through randomized cross-validation techniques, while clinical applications require much more stringent validation based on patient-independent testing. In this study, we show that automatic seizure forecasting can be performed, to some extent, even on independent patients who have never been seen during the training phase, thanks to the implementation of a simple calibration pipeline that can fine-tune deep learning models, even on a single epileptic event recorded from a new patient. We evaluate our calibration procedure using two datasets containing EEG signals recorded from a large cohort of epileptic subjects, demonstrating that the forecast accuracy of deep learning methods can increase on average by more than 20%, and that performance improves systematically in all independent patients. We further show that our calibration procedure works best for deep learning models, but can also be successfully applied to machine learning algorithms based on engineered signal features. Although our method still requires at least one epileptic event per patient to calibrate the forecasting model, we conclude that focusing on realistic validation methods allows to more reliably compare different machine learning approaches for seizure prediction, enabling the implementation of robust and effective forecasting systems that can be used in daily healthcare practice.

Applicability of the International League Against Epilepsy (ILAE) 2022 diagnostic criteria for epilepsy syndromes in children: A retrospective review of 1550 children with epilepsy.

OBJECTIVE: Recently, the ILAE Nosology and Definitions Task Force defined diagnostic criteria for epilepsy syndromes. There is paucity of data on the use of these new diagnostic criteria in children with epilepsy, and how these criteria may lead to changes from previous practice. METHODS: This was a retrospective chart review of data of children attending the epilepsy clinic in a tertiary care children's hospital from January 2011 to January 2023. The clinical details such as age at onset, types of seizures, co-morbidities, and results of EEG, MRI and genetic testing were reviewed. Epilepsy syndrome diagnosis was made as per the ILAE 2022 criteria, and compared with the previous syndrome diagnosis as per records. RESULTS: Data from 1550 children (63 % boys) with epilepsy were analysed, and 55.4 % children were classified to have epilepsy syndromes as per the new ILAE 2022 diagnostic criteria. Application of the new 2022 ILAE diagnostic criteria was associated with a change in name alone in 676 (77.8 %) children. Hundred (11.5 %) children were newly classified under an epilepsy syndrome who had previously remained unclassified. Eleven (1.3 %) children who were previously classified into an epilepsy syndrome could not be classified using the new diagnostic criteria. Eight (0.9 %) were shifted to a new syndromic category. Overall, change in diagnosis occurred in 13.7 (11.5 + 1.3 + 0.9)%. No change in epilepsy syndrome classification/nomenclature occurred in 74 (8.5 %) children. SIGNIFICANCE: The new diagnostic criteria led to an overall change in diagnosis in 13.7 % of children with epilepsy. These criteria will hopefully lead to uniformity in diagnosis of epilepsy syndromes across diverse settings.

Effects of an intervention program to improve mental health and epilepsy care in Madagascar.

BACKGROUND: Despite the high prevalence of mental disorders and epilepsy in low- and middle-income countries, nearly 80% of patients are not treated. In Madagascar, initiatives to improve access to epilepsy and mental health care, including public awareness and training of general practitioners (GPs), were carried out between 2013 and 2018. Our study's main objective was to assess the effectiveness of these initiatives, two to five years post-intervention. METHODS: This quasi-experimental study (intervention vs. control areas) included five surveys assessing: general population's Knowledge Attitudes and Practices (KAP), GPs' KAP , number of epilepsy and mental health consultations at different levels of the healthcare system, diagnostic accuracy, and treatments' availability. OUTCOMES: In the general population, KAP scores were higher in intervention areas for epilepsy (11.4/20 vs. 10.3/20; p = 0.003). For mental disorders, regardless of the area, KAP scores were low, especially for schizophrenia (1.1/20 and 0.1/20). Among GPs, KAP scores were higher in intervention areas for schizophrenia (6.0/10 vs. 4.5/10; p = 0.008) and epilepsy (6.9/10 vs. 6.2/10; p = 0.044). Overall, there was a greater proportion of mental health and epilepsy consultations in intervention areas (4.5% vs 2.3%). Although low, concordance between GPs' and psychiatrists' diagnoses was higher in intervention areas. There was a greater variety of anti-epileptic and psychotropic medications available in intervention areas. INTERPRETATION: This research has helped to better understand the effectiveness of initiatives implemented in Madagascar to improve epilepsy and mental health care and to identify barriers which will need to be addressed. FUNDING: Sanofi Global Health, as part of the Fight Against STigma Program.

Highly Sensitive Near-Field Electrochemical Sensor for In Vivo Monitoring of Respiratory Patterns.

Real-time tracking of respiratory patterns provides noninvasive and quick access for evaluating pathophysiological conditions yet remains challenging due to limited temporal resolution and poor sensitivity to dig out fingerprints of respiratory waveforms. Here, we report an electrochemical sensor for accurately tracing respiratory patterns of small animal models based on the electrochemical impedance mechanism for wireless coupling of a graphdiyne oxide (GYDO)-modified sensing coil chip and a reader coil chip via near-field magnetic induction. In the electrochemical impedance measurement mode, an alternating current is applied through the reader coil chip to perturb proton transport at the GYDO interface of the sensing coil chip. As demonstrated, a high-frequency perturbing condition significantly reduces the interfacial resistance for proton transport by 5 orders of magnitude under 95% relative humidity (RH) and improves the low-humidity responses with a limit of detection down to 0.2% RH, enabling in vivo accurate profiling of respiratory patterns on epileptic rats. The electrochemical impedance coupling system holds great potential for new wireless bioelectronics.

A hybrid 1D CNN-BiLSTM model for epileptic seizure detection using multichannel EEG feature fusion.

Epilepsy, a chronic non-communicable disease is characterized by repeated unprovoked seizures, which are transient episodes of abnormal electrical activity in the brain. While Electroencephalography (EEG) is considered as the gold standard for diagnosis in current clinical practice, manual inspection of EEG is time consuming and biased. This paper presents a novel hybrid 1D CNN-Bi LSTM feature fusion model for automatically detecting seizures. The proposed model leverages spatial features extracted by one dimensional convolutional neural network and temporal features extracted by bi directional long short-term memory network. Ictal and inter ictal data is first acquired from the long multichannel EEG record. The acquired data is segmented and labelled using small fixed windows. Signal features are then extracted from the segments concurrently by the parallel combination of CNN and Bi-LSTM. The spatial and temporal features thus captured are then fused to enhance classification accuracy of model. The approach is validated using benchmark CHB-MIT dataset and 5-fold cross validation which resulted in an average accuracy of 95.90%, with precision 94.78%, F1 score 95.95%. Notably model achieved average sensitivity of 97.18% with false positivity rate at 0.05/hr. The significantly lower false positivity and false negativity rates indicate that the proposed model is a promising tool for detecting seizures in epilepsy patients. The employed parallel path network benefits from memory function of Bi-LSTM and strong feature extraction capabilities of CNN. Moreover, eliminating the need for any domain transformation or additional preprocessing steps, model effectively reduces complexity and enhances efficiency, making it suitable for use by clinicians during the epilepsy diagnostic process.

[Long-term follow-up of adult patients with serial and status course of epileptic seizures]. / Dolgosrochnyi katamnez vzroslykh patsientov s epilepsiei s epilepticheskim statusom ili seriinymi pristupami.

OBJECTIVE: To study the follow-up of adult patients with status epilepticus or a history of serial seizures, assessing the likelihood of achieving long-term remission and identifying predictors of treatment effectiveness. MATERIAL AND METHODS: The study included 280 patients divided into 137 patients with epilepsy with a series of seizures or a history of status epilepticus (group 1) and 143 patients, who had not previously received therapy and did not have a series of seizures or a history of status epilepticus (group 2). A clinical and neurological examination, analysis of medical documentation data, electroencephalography, and MRI were performed. RESULTS: After correction of therapy, remission in patients in group 1 was achieved in 21.9%, improvement in 30%, no effect was observed in 48.1%; in group 2 the indicators were 51%, 28.7%, 20.3%, respectively. The onset of epilepsy in childhood, frequent seizures, and regional epileptiform activity were associated with the lack of treatment effect. CONCLUSION: The results confirm the main role of the clinical examination in determining the prognosis of epilepsy in a particular patient. Currently available instrumental techniques have limited predictive value.

Design and validation of a diagnostic suspicion checklist to differentiate epileptic from psychogenic nonepileptic seizures (PNES-DSC).

OBJECTIVE: Psychogenic non-epileptic seizures (PNES) are complex clinical manifestations and misdiagnosis as status epilepticus remains high, entailing deleterious consequences for patients. Video-electroencephalography (vEEG) remains the gold-standard method for diagnosing PNES. However, time and economic constraints limit access to vEEG, and clinicians lack fast and reliable screening tools to assist in the differential diagnosis with epileptic seizures (ES). This study aimed to design and validate the PNES-DSC, a clinically based PNES diagnostic suspicion checklist with adequate sensitivity (Se) and specificity (Sp) to discriminate PNES from ES. METHODS: A cross-sectional study with 125 patients (n = 104 drug-resistant epilepsy; n = 21 PNES) admitted for a vEEG protocolised study of seizures. A preliminary PNES-DSC (16-item) was designed and used by expert raters blinded to the definitive diagnosis to evaluate the seizure video recordings for each patient. Cohen's kappa coefficient, leave-one-out cross-validation (LOOCV) and balance accuracy (BAC) comprised the main validation analysis. RESULTS: The final PNES-DSC is a 6-item checklist that requires only two to be present to confirm the suspicion of PNES. The LOOCV showed 71.4% BAC (Se = 45.2%; Sp = 97.6%) when the expert rater watched one seizure video recording and 83.4% BAC (Se = 69.6%; Sp = 97.2%) when the expert rater watched two seizure video recordings. CONCLUSION: The PNES-DSC is a straightforward checklist with adequate psychometric properties. With an integrative approach and appropriate patient history, the PNES-DSC can assist clinicians in expediting the final diagnosis of PNES when vEEG is limited. The PNES-DSC can also be used in the absence of patients, allowing clinicians to assess seizure recordings from smartphones.

Automatic Detection of Scalp High-Frequency Oscillations Based on Deep Learning.

Scalp high-frequency oscillations (sHFOs) are a promising non-invasive biomarker of epilepsy. However, the visual marking of sHFOs is a time-consuming and subjective process, existing automatic detectors based on single-dimensional analysis have difficulty with accurately eliminating artifacts and thus do not provide sufficient reliability to meet clinical needs. Therefore, we propose a high-performance sHFOs detector based on a deep learning algorithm. An initial detection module was designed to extract candidate high-frequency oscillations. Then, one-dimensional (1D) and two-dimensional (2D) deep learning models were designed, respectively. Finally, the weighted voting method is used to combine the outputs of the two model. In experiments, the precision, recall, specificity and F1-score were 83.44%, 83.60%, 96.61% and 83.42%, respectively, on average and the kappa coefficient was 80.02%. In addition, the proposed detector showed a stable performance on multi-centre datasets. Our sHFOs detector demonstrated high robustness and generalisation ability, which indicates its potential applicability as a clinical assistance tool. The proposed sHFOs detector achieves an accurate and robust method via deep learning algorithm.

Prognostic Value of Complete Resection of the High-Frequency Oscillation Area in Intracranial EEG: A Systematic Review and Meta-Analysis.

BACKGROUND AND OBJECTIVES: High-frequency oscillations (HFOs; ripples 80-250 Hz; fast ripples [FRs] 250-500 Hz) recorded with intracranial electrodes generated excitement and debate about their potential to localize epileptogenic foci. We performed a systematic review and meta-analysis on the prognostic value of complete resection of the HFOs-area (crHFOs-area) for epilepsy surgical outcome in intracranial EEG (iEEG) accessing multiple subgroups. METHODS: We searched PubMed, Embase, and Web of Science for original research from inception to October 27, 2022. We defined favorable surgical outcome (FSO) as Engel class I, International League Against Epilepsy class 1, or seizure-free status. The prognostic value of crHFOs-area for FSO was assessed by (1) the pooled FSO proportion after crHFOs-area; (2) FSO for crHFOs-area vs without crHFOs-area; and (3) the predictive performance. We defined high combined prognostic value as FSO proportion >80% + FSO crHFOs-area >without crHFOs-area + area under the curve (AUC) >0.75 and examined this for the clinical subgroups (study design, age, diagnostic type, HFOs-identification method, HFOs-rate thresholding, and iEEG state). Temporal lobe epilepsy (TLE) was compared with extra-TLE through dichotomous variable analysis. Individual patient analysis was performed for sex, affected hemisphere, MRI findings, surgery location, and pathology. RESULTS: Of 1,387 studies screened, 31 studies (703 patients) met our eligibility criteria. Twenty-seven studies (602 patients) analyzed FRs and 20 studies (424 patients) ripples. Pooled FSO proportion after crHFOs-area was 81% (95% CI 76%-86%) for FRs and 82% (73%-89%) for ripples. Patients with crHFOs-area achieved more often FSO than those without crHFOs-area (FRs odds ratio [OR] 6.38, 4.03-10.09, p < 0.001; ripples 4.04, 2.32-7.04, p < 0.001). The pooled AUCs were 0.81 (0.77-0.84) for FRs and 0.76 (0.72-0.79) for ripples. Combined prognostic value was high in 10 subgroups: retrospective, children, long-term iEEG, threshold (FRs and ripples) and automated detection and interictal (FRs). FSO after complete resection of FRs-area (crFRs-area) was achieved less often in people with TLE than extra-TLE (OR 0.37, 0.15-0.89, p = 0.006). Individual patient analyses showed that crFRs-area was seen more in patients with FSO with than without MRI lesions (p = 0.02 after multiple correction). DISCUSSION: Complete resection of the brain area with HFOs is associated with good postsurgical outcome. Its prognostic value holds, especially for FRs, for various subgroups. The use of HFOs for extra-TLE patients requires further evidence.

Clinically significant changes in genes and variants associated with epilepsy over time: implications for re-analysis.

Despite the significant advances in understanding the genetic architecture of epilepsy, many patients do not receive a molecular diagnosis after genomic testing. Re-analysing existing genomic data has emerged as a potent method to increase diagnostic yields-providing the benefits of genomic-enabled medicine to more individuals afflicted with a range of different conditions. The primary drivers for these new diagnoses are the discovery of novel gene-disease and variants-disease relationships; however, most decisions to trigger re-analysis are based on the passage of time rather than the accumulation of new knowledge. To explore how our understanding of a specific condition changes and how this impacts re-analysis of genomic data from epilepsy patients, we developed Vigelint. This approach combines the information from PanelApp and ClinVar to characterise how the clinically relevant genes and causative variants available to laboratories change over time, and this approach to five clinical-grade epilepsy panels. Applying the Vigelint pipeline to these panels revealed highly variable patterns in new, clinically relevant knowledge becoming publicly available. This variability indicates that a more dynamic approach to re-analysis may benefit the diagnosis and treatment of epilepsy patients. Moreover, this work suggests that Vigelint can provide empirical data to guide more nuanced, condition-specific approaches to re-analysis.

Feasibility of an Integrated Telemonitoring System for Home-Recorded EEG with Automated AI-Based EEG Analysis.

BACKGROUND: Self-recorded EEG by patients at home might present a viable alternative to inpatient epilepsy evaluations. OBJECTIVES AND METHODS: We developed a novel telemonitoring system comprising seamlessly integrated hard- and software with automated AI-based EEG analysis. RESULTS: The first complete study participation results demonstrate feasibility and clinical utility. CONCLUSION: Our telemonitoring solution potentially improves treatment of patients with epilepsy and moreover might help to better distribute resources in the healthcare system.

[Normal variants with an epileptiform appearance in electroencephalograms. A literature review and clinical implications]. / Variantes normales de aspecto epileptiforme en el electroencefalograma. Revisión de la bibliografía e implicaciones clínicas.

Normal epileptiform-like variants or benign epileptiform variants are a diagnostic challenge in the interpretation of electroencephalograms, which require the knowledge and extensive experience of those responsible for the electroencephalographic report. They include a heterogeneous group of findings, some quite uncommon, initially related to epilepsy and various neurological conditions. Most of them are currently considered variants with no pathological significance, and their over-interpretation usually leads to misdiagnosis and the establishment of unnecessary treatments. Prevalence data are variable and usually come from selected populations, so they are difficult to extrapolate to a healthy population. Studies with invasive electrodes and recent series link some of these variants with epilepsy. We aim to review the characteristics and prevalence of the main benign epileptiform variants and to update their clinical significance.

TITLE: Variantes normales de aspecto epileptiforme en el electroencefalograma. Revisión de la bibliografía e implicaciones clínicas.Las variantes normales de aspecto epileptiforme, o variantes epileptiformes benignas, son un reto diagnóstico en la interpretación de los electroencefalogramas que requiere su conocimiento y una amplia experiencia por parte de los responsables del informe electroencefalográfico. Incluyen un grupo heterogéneo de hallazgos, algunos muy infrecuentes, que inicialmente se relacionaron con epilepsia y patologías neurológicas diversas. En la actualidad, la mayoría se consideran variantes sin significado patológico, y su sobreinterpretación habitualmente acarrea diagnósticos erróneos y tratamientos innecesarios. Los datos de prevalencia de estas variantes son muy diversos y proceden habitualmente de poblaciones seleccionadas, por lo que son difícilmente extrapolables a población sana. No obstante, estudios con electrodos invasivos y series más recientes vuelven a asociar algunas de estas variantes con epilepsia. Nuestro objetivo es revisar las características y la prevalencia de las principales variantes epileptiformes benignas y actualizar su significado clínico.