Search

OBJECTIVE Tailored surgery to extensively resect epileptogenic lesions using intraoperative electrocorticography (ioECoG) may improve seizure outcomes. However, resection of large areas is associated with decreased memory function postoperatively. The authors assessed whether ioECoG could provide useful information on how to minimize the focus resection and obtain better seizure outcomes without memory deterioration. They examined the postoperative seizure-free period and memory alteration in a retrospective cohort of patients with mesial temporal lobe epilepsy (TLE) due to hippocampal sclerosis (HS) in whom the extent of removal was determined using ioECoG findings. METHODS The authors enrolled 82 patients with TLE associated with HS who were treated surgically. Transsylvian amygdalohippocampectomy was indicated as the first step. When visual inspection identified interictal epileptic discharges from the lateral temporal lobe on ioECoG, anterior temporal lobectomy (ATL) was eventually performed. The patients were divided into the selective amygdalohippocampectomy (SA, n = 40) and ATL (n = 42) groups. Postoperative seizure outcomes were assessed at 1, 2, 3, 5, and 7 years postoperatively using the International League Against Epilepsy classification. The Kaplan-Meier survival analysis was applied to evaluate the period of seizure recurrence between the SA and ATL groups. Factors attributed to seizure recurrence were analyzed using the Cox proportional hazards model, and they were as follows: epileptic focal laterality; age at seizure onset (< 10 or ≥ 10 years old); seizure frequency (more than weekly or less than weekly seizures); history of focal to bilateral tonic-clonic seizure; infectious etiology; and surgical procedure. The Wechsler Memory Scale–Revised was used to evaluate memory function pre- and postoperatively. RESULTS Seizure outcomes were significantly worse in the SA group than in the ATL group at 2 years postoperatively (p = 0.045). The International League Against Epilepsy class 1 outcomes at 7 years postoperatively in the SA and ATL groups were 63% and 81%, respectively. Kaplan-Meier analysis showed that seizure recurred significantly earlier in the SA group than in the ATL group (p = 0.031). The 2-way ANOVA analysis was used to compare the SA and ATL groups in each memory category, and revealed that there was no significant difference regardless of the side of surgery. CONCLUSIONS Visual assessment of ioECoG cannot be used as an indicator to minimize epileptic focus resection in patients with TLE associated with HS. ATL is more effective in obtaining seizure-free outcomes; however, both ATL and SA can preserve memory function.

The impact of the coronavirus disease 2019 (COVID-19) pandemic on epilepsy care across Japan was investigated by conducting a multicenter retrospective cohort study.This study included monthly data on the frequency of (1) visits by outpatients with epilepsy, (2) outpatient electroencephalography (EEG) studies, (3) telemedicine for epilepsy, (4) admissions for epilepsy, (5) EEG monitoring, and (6) epilepsy surgery in epilepsy centers and clinics across Japan between January 2019 and December 2020. We defined the primary outcome as epilepsy-center-specific monthly data divided by the 12-month average in 2019 for each facility. We determined whether the COVID-19 pandemic-related factors (such as year [2019 or 2020], COVID-19 cases in each prefecture in the previous month, and the state of emergency) were independently associated with these outcomes.In 2020, the frequency of outpatient EEG studies (-10.7%, p0.001) and cases with telemedicine (+2,608%, p=0.031) were affected. The number of COVID-19 cases was an independent associated factor for epilepsy admission (-3.75*10We demonstrated the significant impact that the COVID-19 pandemic had on epilepsy care. These results support those of previous studies and clarify the effect size of each pandemic-related factor on epilepsy care.

Electroencephalogram (EEG) is one of most effective clinical diagnosis modalities for the localization of epileptic focus. Most current AI solutions use this modality to analyze the EEG signals in an automated manner to identify the epileptic seizure focus. To develop AI system for identifying the epileptic focus, there are many recently-published AI solutions based on biomarkers or statistic features that utilize interictal EEGs. In this review, we survey these solutions and find that they can be divided into three main categories: (i) those that use of biomarkers in EEG signals, including high-frequency oscillation, phase-amplitude coupling, and interictal epileptiform discharges, (ii) others that utilize feature-extraction methods, and (iii) solutions based upon neural networks (an end-to-end approach). We provide a detailed description of seizure focus with clinical diagnosis methods, a summary of the public datasets that seek to reduce the research gap in epilepsy, recent novel performance evaluation criteria used to evaluate the AI systems, and guidelines on when and how to use them. This review also suggests a number of future research challenges that must be overcome in order to design more efficient computer-aided solutions to epilepsy focus detection.

To examine the current medical and psychosocial status of patients with epilepsy, aiming to facilitate appropriate application of the Intractable/Rare Diseases Act of Japan.By analysing the cross-sectional data of patients registered in the tertiary hospital-based Epilepsy Syndrome Registry of Japan, we investigated the proportion of patients who met the severity criteria as defined by the Act (seizure frequency of at least once a month, or presence of intellectual/neurological/psychiatric symptoms, or both) and whether there are candidate syndrome/diseases to be added to the existing list in the Act.In total, 2,209 patients were registered. After excluding self-limited/idiopathic epilepsies, 1,851 of 2,110 patients (87.7%) met the severity criteria. The patients were classified into eight main epilepsy syndromes (594 patients), 20 groups based on aetiology (1,078 patients), and three groups without known aetiology (427 patients). Most of the groups classified by syndrome or aetiology had high proportions of patients satisfying the severity criteria (90%), but some groups had relatively low proportions (80%) resulting from favourable outcome of surgical therapy. Several small groups with known syndrome/aetiology await detailed analysis based on a sufficiently large enough number of patients registered, some of whom may potentially be added to the list of the Act.The registry provides data to examine the usefulness of the severity criteria and list of diseases that are operationally defined by the Act. Most epilepsy patients with various syndromes/diseases and aetiology groups are covered by the Act but some are not, and the list of designated syndromes/diseases should be complemented by further amendments, as suggested by future research.

Objective. Accurate detection of epileptic seizures using electroencephalogram (EEG) data is essential for epilepsy diagnosis, but the visual diagnostic process for clinical experts is a time-consuming task. To improve efficiency, some seizure detection methods have been proposed. Regardless of traditional or machine learning methods, the results identify only seizures and non-seizures. Our goal is not only to detect seizures but also to explain the basis for detection and provide reference information to clinical experts. Approach. In this study, we follow the visual diagnosis mechanism used by clinical experts that directly processes plotted EEG image data and apply some commonly used models of LeNet, VGG, deep residual network (ResNet), and vision transformer (ViT) to the EEG image classification task. Before using these models, we propose a data augmentation method using random channel ordering (RCO), which adjusts the channel order to generate new images. The Gradient-weighted class activation mapping (Grad-CAM) and attention layer methods are used to interpret the models. Main results. The RCO method can balance the dataset in seizure and non-seizure classes. The models achieved good performance in the seizure detection task. Moreover, the Grad-CAM and attention layer methods explained the detection basis of the model very well and calculate a value that measures the seizure degree. Significance. Processing EEG data in the form of images can flexibility to use a variety of machine learning models. The imbalance problem that exists widely in clinical practice is well solved by the RCO method. Since the method follows the visual diagnosis mechanism of clinical experts, the model interpretation results can be presented to clinical experts intuitively, and the quantitative information provided by the model is also a good diagnostic reference.

Automatic seizure onset zone (SOZ) localization using interictal electrocorticogram (ECoG) improves the diagnosis and treatment of patients with medically refractory epilepsy. This study aimed to investigate the characteristics of phase-amplitude coupling (PAC) extracted from interictal ECoG and the feasibility of PAC serving as a promising biomarker for SOZ identification. We employed the mean vector length modulation index approach on the 20-s ECoG window to calculate PAC features between low-frequency rhythms (0.5–24 Hz) and high frequency oscillations (HFOs) (80–560 Hz). We used statistical measures to test the significant difference in PAC between the SOZ and non-seizure onset zone (NSOZ). To overcome the drawback of handcraft feature engineering, we established novel machine learning models to learn automatically the characteristics of the obtained PAC features and classify them to identify the SOZ. Besides, to handle imbalanced dataset classification, we introduced novel feature-wise/class-wise re-weighting strategies in conjunction with classifiers. In addition, we proposed a time-series nest cross-validation to provide more accurate and unbiased evaluations for this model. Seven patients with focal cortical dysplasia were included in this study. The experiment results not only showed that a significant coupling at band pairs of slow waves and HFOs exists in the SOZ when compared with the NSOZ, but also indicated the effectiveness of the PAC features and the proposed models in achieving better classification performance .

Purpose The removal of the bottom of sulcus dysplasia (BOSD) often includes the gyral crown; however, this method has been controversial. We hypothesized that the epileptogenic zone of the BOSD does not include the gyral crown. To reveal the depth and extent of the epileptogenic zone of the BOSD, we applied the two electrophysiological modalities: (1) the occurrence rate (OR) of high-frequency oscillations (HFOs) and (2) modulation index (MI), reflecting the strength of phase-amplitude coupling between HFOs and slow oscillations. Methods We investigated the ripples [80-200 Hz] and fast ripples [200-300 Hz]) in HFOs and MI (HFOs [80-300 Hz] and slow oscillations [3-4 Hz]). We opened the sulcus at the BOSD and implanted the subdural electrodes directly over the MRI visible lesion. All patients (n = 3) underwent lesionectomy and the gyral crown was preserved. Results Pathological findings demonstrated focal cortical dysplasia type IIb and seizure freedom was achieved. The OR of the HFOs was not significantly different between the BOSD and the gyral crown. In contrast, the MI between HFOs and slow oscillations in the BOSD was significantly higher than that in the gyral crown. Conclusion High MI values distinguished the epileptogenic BOSD from the non-epileptogenic gyral crowns. MI could be a more informative biomarker of epileptogenicity than the OR of HFOs in a subset of patients with the BOSD.

After establishing latent infection, some viruses can be reactivated by the alteration of host immunological conditions. First, we reviewed viruses that can cause neuronal damage by reactivation. Then we focused on the herpes simplex virus (HSV). The reactivation leads to neuronal damages through two possible mechanisms; "reactivation of a latent herpes virus" by which viruses can cause direct virus neurotoxicity, and "post-infectious immune inflammatory response" by which a focal reactivation of HSV leads to an inflammatory reaction. The former is radiologically characterized by cortical lesions, the latter is characterized by subcortical white matter lesions. We experienced a female, who underwent the right posterior quadrantectomy and then developed recurrent herpes encephalitis caused by herpes simplex reactivation, which pathologically demonstrated inflammation in the white matter, suggesting a post-infectious immune inflammatory response. The patient was successfully treated with immunosuppressants. The reactivation of the HSV is extremely rare in Japan. Neurologists should recognize this condition because this disorder will increase as epilepsy surgery gains more popularity.

Epilepsy is a chronic disorder caused by excessive electrical discharges. Currently, clinical experts identify the seizure onset zone (SOZ) channel through visual judgment based on long-time intracranial electroencephalogram (iEEG), which is a very time-consuming, difficult and experience-based task. Therefore, there is a need for high-accuracy diagnostic aids to reduce the workload of clinical experts. In this article, we propose a method in which, the iEEG is split into the 20-s segment and for each patient, we ask clinical experts to label a part of the data, which is used to train a model and classify the remaining iEEG data. In recent years, machine learning methods have been successfully applied to solve some medical problems. Filtering, entropy and short-time Fourier transform (STFT) are used for extracting features. We compare them to wavelet transform (WT), empirical mode decomposition (EMD) and other traditional methods with the aim of obtaining the best possible discriminating features. Finally, we look for their medical interpretation, which is important for clinical experts. We achieve high-performance results for SOZ and non-SOZ data classification by using the labeled iEEG data and support vector machine (SVM), fully connected neural network (FCNN) and convolutional neural network (CNN) as classification models. In addition, we introduce the positive unlabeled (PU) learning to further reduce the workload of clinical experts. By using PU learning, we can learn a binary classifier with a small amount of labeled data and a large amount of unlabeled data. This can greatly reduce the amount and difficulty of annotation work by clinical experts. All together, we show that using 105 minutes of labeled data we achieve a classification result of 91.46% on average for multiple patients.

Background Individuals with Sturge-Weber syndrome (SWS) often expereince intractable epilepsy and cognitive decline. We hypothesized that the extent of the leptomeningeal capillary malformation (LCM) may correlate with the severity of neurological impairment due to SWS. We tested the hypothesis in a cross-sectional study of seizure severity and electroencephalographic (EEG) findings and a retrospective cohort study for surgical indications related to the extent of the LCM. Methods We enrolled 112 patients and classified them according to LCM distribution: (1) bilateral, (2) hemispheric, (3) multilobar, and (4) single lobe. Age at seizure onset, seizure semiology and frequency, and EEG findings were compared. Surgical indications were evaluated for each group by Fisher exact test, and predictors for surgery were evaluated by univariate and multivariate analyses. Therapeutic efficacy was evaluated by the SWS-Neurological Score (SWS-NS). Results The bilateral and hemispheric groups had early seizure onset (4.0 months old and 3.0 months old), frequent seizures (88.9% and 80.6% had more than one per month), focal-to-bilateral tonic-clonic seizures (88.9% and 74.2%), and status epilepticus (100% and 87.1%). The groups’ EEG findings did not differ substantially. Surgical indications were present in 77.8% of the bilateral, 88.1% of the hemispheric, and 46.8% of the multilobar groups. Seizure more than once per month was a predictor of surgical treatment. Seizure subscore improved postoperatively in the hemispheric and multilobar groups. Even after surgical treatment, the bilateral and hemispheric groups exhibited higher SWS-NSs than members of the other groups. Conclusion Our study demonstrated a strong association between extensive LCM and epilepsy severity. Surgical intervention improved seizure outcome in patients with SWS with large LCMs.

Late relapse of herpes simplex encephalitis (HSE) is defined as the recurrence of HSE more than 3 months after the initial exposure. The postoperative diagnosis of HSE following neurosurgery is complicated because the clinical presentation can mimic other common complications of neurosurgery. Cerebrospinal fluid polymerase chain reactions (CSF-PCR) is the gold standard for the diagnosis of HSE. We describe a case of late HSE relapse after epilepsy surgery in a patient who required a brain biopsy due to repeated negative CSF-PCR results. A 38-year-old woman had a history of HSE from the age of 3 years. She had intractable epilepsy from the age of 20 years and underwent right posterior quadrant disconnection (PQD) at the age of 38 years. Postoperatively, she had a right hemispheric intracerebral hemorrhage (ICH) and her consciousness was gradually worsening. Her consciousness improved after removal of the ICH. However, her consciousness gradually deteriorated again. Fluid-attenuated inversion recovery (FLAIR) revealed bilateral hyperintensity in the frontal lobes, including the white matter. CSF-PCR for herpes simplex virus (HSV) was performed twice, but yielded negative results. We performed a brain biopsy to target FLAIR hyperintensity in the right frontal lobe. PCR of the brain specimen was positive for HSV. Her consciousness improved with acyclovir, methylprednisolone, and cyclophosphamide. To our knowledge, this is a case of HSE induced by epilepsy surgery which had the longest duration until relapse after the initial HSE episode. A brain biopsy can be used to confirm the diagnosis of suspected HSE when CSF-PCR results are negative.

ObjectiveBecause of the lack of highly skilled experts, automated technologies that support electroencephalogram (EEG)-based in epilepsy diagnosis are advancing. Deep convolutional neural network-based models have been used successfully for detecting epileptic spikes, one of the biomarkers, from EEG. However, a sizeable number of supervised EEG records are required for training.ApproachThis study introduces the Satelight model, which uses the self-attention (SA) mechanism. The model was trained using a clinical EEG dataset labeled by five specialists, including 16,008 epileptic spikes and 15,478 artifacts from 50 children. The SA mechanism is expected to reduce the number of parameters and efficiently extract features from a small amount of EEG data. To validate the effectiveness, we compared various spike detection approaches with the clinical EEG data.Main resultsThe experimental results showed that the proposed method detected epileptic spikes more effectively than other models (accuracy = 0.876 and false positive rate = 0.133).SignificanceThe proposed model had only one-tenth the number of parameters as the other effective model, despite having such a high detection performance. Further exploration of the hidden parameters revealed that the model automatically attended to the EEG’s characteristic waveform locations of interest.

Epileptic seizure is the common symptom associated with lipomas in the Sylvian fissure (Sylvian lipomas). Removal of these lipomas carries risks of hemorrhage and brain damage. We report a surgical strategy of not removing the lipoma in a case of intractable temporal lobe epilepsy associated with Sylvian lipoma. We performed anterior temporal lobectomy with preservation of the pia mater of the Sylvian fissure and achieved seizure freedom. Focal cortical dysplasia type 1 of the epileptic neocortex adjacent to the Sylvian lipoma was pathologically diagnosed. We recommend our surgical procedure in similar cases to avoid complications and achieve adequate seizure control.

Presurgical investigations for categorizing focal patterns are crucial, leading to localization and surgical removal of the epileptic focus. This paper presents a machine learning approach using information theoretic features extracted from high-frequency subbands to detect the epileptic focus from interictal intracranial electroencephalogram (iEEG). It is known that high-frequency subbands (>80 Hz) include important biomarkers such as high-frequency oscillations (HFOs) for identifying epileptic focus commonly referred to as the seizure onset zone (SOZ). In this analysis, the multi-channel interictal iEEG signals were splitted into segments and each segment was decomposed into multiple high-frequency subbands. The different types of entropy were calculated for each of the subbands and the sparse linear discriminant analysis (sLDA) was applied to select the prominent entropy features. Due to the imbalance of SOZ and non-SOZ channels in iEEG data, the use of machine learning techniques is always tricky. To deal with the imbalanced learning problem, an adaptive synthetic oversampling approach (ADASYN) with radial basis function kernel-based SVM was used to detect the focal segments. Finally, the epileptic focus was identified based on detection of focal segments on SOZ and non-SOZ channels. Eight patients were examined to observe the efficiency of the automatic detector. The experimental results and statistical tests indicate that the proposed automatic detector can identify the epileptic focus accurately and efficiently.

We assessed the diagnostic utility of the occurrence rate of high-frequency oscillations and modulation index (MI) from intraoperative electrocorticography (ioECoG) in determining the extent of epileptogenicity in mesial temporal lobe epilepsy (TLE) with hippocampal sclerosis (HS).We enrolled 17 patients who underwent selective amygdalohippocampectomy (SelAH) for TLE due to HS. We analyzed the occurrence rate of ripples (80-200 Hz) and fast ripples (200-300 Hz); and MI between ripples and 3-4 Hz (MIIn the poor seizure outcome group, an increase in the occurrence rate of ripples was seen in the hippocampus and LTL pre-SelAH and the LTL post-SelAH. The MIHigh occurrence rate of ripples and MI

OBJECTIVE To identify the risk factors for psychological distress in electroencephalography (EEG) technicians during the coronavirus disease 2019 (COVID-19) pandemic. METHOD In this national-level cross-sectional survey initiated by Japan Young Epilepsy Section (YES-Japan), which is a national chapter of The Young Epilepsy Section of the International League Against Epilepsy (ILAE-YES), a questionnaire was administered to 173 technicians engaged in EEG at four clinics specializing in epilepsy care and 20 hospitals accredited as (quasi-) epilepsy centers or epilepsy training facilities in Japan from March 1 to April 30, 2021. We collected data on participants' profiles, information about work, and psychological distress outcome measurements, such as the K-6 and Tokyo Metropolitan Distress Scale for Pandemic (TMDP). Linear regression analysis was used to identify the risk factors for psychological distress. Factors that were significantly associated with psychological distress in the univariate analysis were subjected to multivariate analysis. RESULTS Among the 142 respondents (response rate: 82%), 128 were included in the final analysis. As many as 35.2% of EEG technicians have been under psychological distress. In multivariate linear regression analysis for K-6, female sex, examination for patients (suspected) with COVID-19, and change in salary or bonus were independent associated factors for psychological distress. Contrastingly, in multivariate linear regression analysis for TMDP, female sex, presence of cohabitants who had to be separated from the respondent due to this pandemic, and change in salary or bonus were independent associated factors for psychological distress. CONCLUSION We successfully identified the risk factors associated with psychological distress in EEG technicians during the COVID-19 pandemic. Our results may help in understanding the psychological stress in EEG technicians during the COVID-19 pandemic and improving the work environment, which is necessary to maintain the mental health of EEG technicians.

Objective To unveil current medical and psychosocial conditions of patients with West syndrome in Japan. Methods A cross-sectional analysis was performed in patients with West syndrome registered in the Rare Epilepsy Syndrome Registry (RES-R) of Japan. Furthermore, new-onset patients registered in the RES-R were observed prospectively and their outcomes after one and two years of follow-up were compared with data at onset. Results For the cross-sectional study, 303 patients with West syndrome were included. Seizures (such as spasms, tonic seizures and focal seizures) occurred daily in 69.3% of the patients at registration. Seizure frequency of less than one per year was observed in cases of unknown etiology (22.6%), genetic etiology (23.8%) and malformation of cortical development (MCD; 19.1%). Neurological findings were absent in 37.0%, but a high rate of abnormality was seen in patients with Aicardi syndrome, hypoxic-ischemic encephalopathy (HIE), genetic etiology and MCD other than focal cortical dysplasia, accompanied by a >50% rate of bedridden patients. Abnormal EEG was found in 96.7%, and CT/MRI was abnormal in 62.7%. Treatments included antiepileptic drug therapy (94.3%), hormonal therapy (72.6%), diet therapy (8.3%) and surgery (15.8%). Intellectual/developmental delay was present in 88.4%, and was more severe in patients with Aicardi syndrome, genetic etiology and HIE. Autism spectrum disorder was found in 13.5%. For the longitudinal study, 27 new-onset West syndrome patients were included. The follow-up study revealed improved seizure status after two years in 66.7%, but worsened developmental status in 55.6%, with overall improvement in 51.9%. Significance The study reveals the challenging neurological, physical and developmental aspects, as well as intractable seizures, in patients with West syndrome. More than a half of the children showed developmental delay after onset, even though seizures were reduced during the course of the disease.

Quantitative magnetic resonance imaging (MRI) with multislice, multi-echo, and multi-delay acquisition enables simultaneous quantification of R1 and R2 relaxation rates, proton density, and the B1 field in a single acquisition, and requires only about 6 minutes for full-head coverage. Using dedicated SyMRI software, radiologists can generate any contrast-weighted image by manipulating the acquisition parameters, including repetition time, echo time, and inversion time. Moreover, automatic brain tissue segmentation, volumetry, and myelin measurement can also be performed. Using the SyMRI approach, a shorter scan time, an objective examination, and personalized MR imaging parameters can be obtained in daily clinical pediatric imaging. Here we summarize and review the use of SyMRI in imaging of the pediatric brain, including the basic principles of MR quantification along with its features, clinical applications, and limitations.

Background Patients with idiopathic normal-pressure hydrocephalus (iNPH) are typically older adults with multiple comorbidities that are associated with a reduction in the efficacy of iNPH treatment via cerebrospinal fluid (CSF) shunt placement. Objective The present study aimed to investigate the effectiveness of CSF shunt for iNPH using data from a nationwide epidemiological survey in Japan. Methods We examined 1,423 patients (581 women) aged ≥60 years (median age [25%-75%]: 77 [73-80] years) who were diagnosed with iNPH following a hospital visit in 2012. Patients who experienced an improvement of at least one modified Rankin Scale (mRS) grade after the CSF shunt were classified as "improvement" while the remaining patients were classified as "non-improvement." The efficacy of the shunt intervention (n = 842) was analyzed using a binomial logistic regression analysis. Results An analysis of risk factors associated with shunt placement in patients with mRS grade 2 revealed an association between comorbid chronic ischemic lesions (odds ratio [OR], 2.28; 95% confidence interval [CI], 1.11-4.67; p = 0.025) and cervical spondylosis (OR, 3.62; 95% CI, 1.15-11.34; p = 0.027). Patients with mRS grade 3 at study entry had an association with comorbid Alzheimer's disease (OR, 3.02; 95% CI, 1.44-6.31; p = 0.003). Conclusions The results presented here showed that any age-related risk is minimal and should not be cause for rejection of surgical treatment options. Clinical decisions regarding CSF shunt should be individualized to each patient, with adequate consideration of the relative risks and benefits, including maximizing a healthy life expectancy.

In athletes, long-term intensive training has been shown to increase unparalleled athletic ability and might induce brain plasticity. We evaluated the structural connectome of world-class gymnasts (WCGs), as mapped by diffusion-weighted magnetic resonance imaging probabilistic tractography and a multishell, multitissue constrained spherical deconvolution method to increase the precision of tractography at the tissue interfaces. The connectome was mapped in 10 Japanese male WCGs and in 10 age-matched male controls. Network-based statistic identified subnetworks with increased connectivity density in WCGs, involving the sensorimotor, default mode, attentional, visual, and limbic areas. It also revealed a significant association between the structural connectivity of some brain structures with functions closely related to the gymnastic skills and the D-score, which is used as an index of the gymnasts' specific physical abilities for each apparatus. Furthermore, graph theory analysis demonstrated the characteristics of brain anatomical topology in the WCGs. They displayed significantly increased global connectivity strength with decreased characteristic path length at the global level and higher nodal strength and degree in the sensorimotor, default mode, attention, and limbic/subcortical areas at the local level as compared with controls. Together, these findings extend the current understanding of neural mechanisms that distinguish WCGs from controls and suggest brain anatomical network plasticity in WCGs resulting from long-term intensive training. Future studies should assess the contribution of genetic or early-life environmental factors in the brain network organization of WCGs. Furthermore, the indices of brain topology (i.e., connection density and graph theory indices) could become markers for the objective evaluation of gymnastic performance.

The design of a computer-aided system for identifying the seizure onset zone (SOZ) from interictal and ictal electroencephalograms (EEGs) is desired by epileptologists. This study aims to introduce the statistical features of high-frequency components (HFCs) in interictal intracranial electroencephalograms (iEEGs) to identify the possible seizure onset zone (SOZ) channels. It is known that the activity of HFCs in interictal iEEGs, including ripple and fast ripple bands, is associated with epileptic seizures. This paper proposes to decompose multi-channel interictal iEEG signals into a number of subbands. For every 20 s segment, twelve features are computed from each subband. A mutual information (MI)-based method with grid search was applied to select the most prominent bands and features. A gradient-boosting decision tree-based algorithm called LightGBM was used to score each segment of the channels and these were averaged together to achieve a final score for each channel. The possible SOZ channels were localized based on the higher value channels. The experimental results with eleven epilepsy patients were tested to observe the efficiency of the proposed design compared to the state-of-the-art methods.

To cope with the lack of highly skilled professionals, machine learning with proper signal processing is key for establishing automated diagnostic-aid technologies with which to conduct epileptic electroencephalogram (EEG) testing. In particular, frequency filtering with the appropriate passbands is essential for enhancing the biomarkers—such as epileptic spike waves—that are noted in the EEG. This paper introduces a novel class of neural networks (NNs) that have a bank of linear-phase finite impulse response filters at the first layer as a preprocessor that can behave as bandpass filters that extract biomarkers without destroying waveforms because of a linear-phase condition. Besides, the parameters of the filters are also data-driven. The proposed NNs were trained with a large amount of clinical EEG data, including 15,833 epileptic spike waveforms recorded from 50 patients, and their labels were annotated by specialists. In the experiments, we compared three scenarios for the first layer: no preprocessing, discrete wavelet transform, and the proposed data-driven filters. The experimental results show that the trained data-driven filter bank with supervised learning behaves like multiple bandpass filters. In particular, the trained filter passed a frequency band of approximately 10–30 Hz. Moreover, the proposed method detected epileptic spikes, with the area under the receiver operating characteristic curve of 0.967 in the mean of 50 intersubject validations.