EEG classification approach based on the extreme learning machine and wavelet transform

Automatic detection and classification of electroencephalogram (EEG) epileptic activity aid diagnosis and relieve the heavy workload of doctors. This article presents a new EEG classification approach based on the extreme learning machine (ELM) and wavelet transform (WT). First, the WT is used to extract useful features when certain scales cover abnormal components of the EEG. Second, the ELM algorithm is used to train a single hidden layer of feedforward neural network (SLFN) features. Finally, the SLFN is tested with interictal and ictal EEGs. The experiments demonstrated that the proposed approach achieved a satisfactory classification rate of 99.25% for interictal and ictal EEGs.     

EEG non-linear feature extraction using correlation dimension and Hurst exponent

In this work, we evaluated the differences between epileptic electroencephalogram (EEG) and interictal EEG by computing some non-linear features. Correlation dimension (CD) and Hurst exponent (H) were calculated for 100 segments of epileptic EEG and 100 segments of interictal EEG. A comparison was made between epileptic EEG and interictal EEG in those non-linear parameters. Results show that the mean values of CD are 2.64 for epileptic EEG and 4.55 for interictal EEG. We also calculated approximate entropy (ApEn) of those EEG signals. The mean values of ApEn are 0.90 for epileptic EEG and 4.55 for interictal EEG. The values of CD and ApEn of epileptic EEG are generally lower than those of interictal EEG, indicating less complexity of EEG signals during seizures. The mean values of Hurst exponent are 0.19 for epileptic EEG and 0.29 for interictal EEG. Hurst exponents for epileptic EEG and interictal EEG are both <0.5. This indicates that both epileptic and interictal EEGs show long-range anticorrelation. The value of Hurst exponent of epileptic EEG signals is lower than that of interictal EEG signals, showing that the degree of anticorrelation of epileptic EEG signals is larger than that of interictal EEG. Hence, the non-linear parameters such as CD and Hurst exponent can help interpret epileptic and interictal EEGs and their neurodynamics.     

EEG non-linear feature extraction using correlation dimension and Hurst exponent

Abstract

In this work, we evaluated the differences between epileptic electroencephalogram (EEG) and interictal EEG by computing some non-linear features. Correlation dimension (CD) and Hurst exponent (H) were calculated for 100 segments of epileptic EEG and 100 segments of interictal EEG. A comparison was made between epileptic EEG and interictal EEG in those non-linear parameters. Results show that the mean values of CD are 2·64 for epileptic EEG and 4·55 for interictal EEG. We also calculated approximate entropy (ApEn) of those EEG signals. The mean values of ApEn are 0·90 for epileptic EEG and 4·55 for interictal EEG. The values of CD and ApEn of epileptic EEG are generally lower than those of interictal EEG, indicating less complexity of EEG signals during seizures. The mean values of Hurst exponent are 0·19 for epileptic EEG and 0·29 for interictal EEG. Hurst exponents for epileptic EEG and interictal EEG are both <0·5. This indicates that both epileptic and interictal EEGs show long-range anticorrelation. The value of Hurst exponent of epileptic EEG signals is lower than that of interictal EEG signals, showing that the degree of anticorrelation of epileptic EEG signals is larger than that of interictal EEG. Hence, the non-linear parameters such as CD and Hurst exponent can help interpret epileptic and interictal EEGs and their neurodynamics.     

Independent component analysis of ictal EEG in medial temporal lobe epilepsy

PURPOSE: Application of independent component analysis (ICA) to interictal EEGs and to event-related potentials has helped noise reduction and source localization. However, ICA has not been used for the analysis of ictal EEGs in partial seizures. In this study, we applied ICA to the ictal EEGs of patients with medial temporal lobe epilepsy (TLE) and investigated whether ictal components can be separated and whether they indicate correct lateralization. METHODS: Twenty-four EEGs from medial TLE patients were analyzed with the extended ICA algorithm. Among the resultant 20 components in each EEG, we selected components with an ictal nature and reviewed their corresponding topographic maps for the lateralization. We then applied quantitative methods for the verification of increased quality of the reconstructed EEGs. RESULTS: All ictal EEGs were successfully decomposed into one or more ictal components and nonictal components. After EEG reconstruction with exclusion of artifacts, the lateralizing power of the ictal EEG was increased from 75 to 96%. CONCLUSIONS: ICA can separate successfully the manifold components of ictal rhythms and can improve EEG quality.     

Non-linear analysis of intracranial human EEG in temporal lobe epilepsy.

OBJECTIVE: Intracranial EEG recordings from patients suffering from medically intractable temporal lobe epilepsy were analyzed with the aim of characterizing the dynamics of EEG epochs recorded before and during a seizure and comparing the classification of the EEG epochs on the basis of visual inspection to the results of the numerical analysis. METHODS: The stationarity of the selected EEGs was assessed qualitatively. The coarse-grained correlation dimension and coarse-grained correlation entropy were used for the non-linear characterization of the EEG epochs. RESULTS: High-pass filtering was necessary in order to make the majority of the epochs appear stationarity beyond a time scale of about 2 s. It was found that the dimension of the ictal EEGs decreased with respect to the epochs containing ongoing (interictal) activity. The entropy of the ictal recordings however increased. A scaling of the entropy was applied and it was found that the scaled entropy of the ictal EEG decreased, consistent with the increased regularity of the ictal EEG. The coarse-grained quantities discriminated well between EEG epochs recorded prior to and during seizures at locations displaying ictal activity and classification improved by including the linear autocorrelation time in the analysis. CONCLUSIONS: It is concluded that ictal and non-ictal EEG can be well distinguished on the basis of non-linear analysis. The results are in good agreement with the visual analysis.     

头皮脑电图在功能性大脑半球切除术中的应用价值

目的探讨头皮脑电图(EEG)在功能性大脑半球切除术中的应用价值。方法回顾性分析8例药物难治性癫痫行功能性大脑半球切除术的患者,分析病因与术后转归相关性,着重分析术前头皮EEG背景活动、间歇期放电、发作起始及术后EEG特征。结果 8例患者以获得性病因为主。术前头皮EEG背景活动多表现为慢波增多,患侧显著;发作间歇期双侧半球呈现多脑区放电特征;发作起始EEG改变多为弥漫性改变;术后复查EEG,1个月内以双侧放电居多,半年以上复查以患侧放电居多。结论头皮EEG作为一种简便易行的检查手段,在功能性大脑半球切除术术前评估、术后随访中起到重要作用。     

Utility of the Scalp-Recorded Ictal EEG in Childhood Epilepsy

PURPOSE: To evaluate the usefulness of the scalp-recorded ictal EEGs in diagnosing childhood epilepsy. METHODS: We analyzed the ictal EEGs of 259 seizures in 183 patients who visited the department of child neurology, Okayama University Medical School, during the past 6 years. RESULTS: We divided all seizures into the following four categories, according to the diagnostic usefulness of ictal EEGs in determining the seizure type: 1. (a) Ictal EEGs confirmed the diagnosis of the seizure type based on seizure symptoms (101 seizures); (b) Ictal EEGs aided in the classification of the seizure type based on the seizure symptoms (101 seizures); (c) Ictal EEGs corrected errors in the classification (37 seizures); and (d) Ictal EEGs revealed previously unreported/undocumented seizure type (20 seizures). 2. Of the 37 misdiagnosed seizures (group C), 11 were nonepileptic seizures misdiagnosed as epileptic seizures, eight were complex partial seizures (CPS) misdiagnosed as the other seizure types, and 10 were other seizure types misdiagnosed as CPSs. 3. Of the 20 previously unreported/undocumented seizures (group D), nine were myoclonic seizures, five were absence seizures, five were CPS, and one was tonic spasms. 4. Seventy-two patients had CPS. Among them, 11 patients showed no epileptic spikes in their interictal EEG recordings. Therefore, ictal recordings confirmed the diagnosis of epilepsy. CONCLUSIONS: Ictal EEG recording is a very useful diagnostic tool not only for determining seizure types, but also for uncovering the existence of the unsuspected seizure types. It supplies the physician with useful information for the classification and the treatment of epilepsy. In particular, ictal EEGs are useful in diagnosing patients with CPS.     

Is ictal recording mandatory in temporal lobe epilepsy? Not when the interictal electroencephalogram and hippocampal atrophy coincide

OBJECTIVE: To investigate the concordance between scalp electroencephalogram (EEG) lateralization and side of hippocampal atrophy in patients with temporal lobe epilepsy (TLE). METHODS: We studied 184 consecutive patients with TLE without lesions other than those compatible with mesial temporal sclerosis. In this study, we studied specifically hippocampal atrophy and the results of scalp EEG investigation. Patients were classified according to the localization of interictal epileptiform discharges as unilateral, bilateral asymmetric, and bilateral symmetric. The EEG seizure onsets were also classified separately as unilateral, bilateral asymmetric, and bilateral symmetric. The hippocampal atrophy was determined by volumetric measurements using high-resolution magnetic resonance imaging (MRIVol). RESULTS: Only 3% of patients had discordance between the ictal and interictal EEG lateralizations; however, none of these had unilateral interictal EEG abnormalities. Interictal EEGs were considered unilateral in 62.0% of patients, bilateral asymmetric in 31.5%, and bilateral symmetric in 6.5%. Ictal EEGs were considered unilateral in 63.5% of patients, bilateral asymmetric in 30.0%, and bilateral symmetric in 6.5%. The MRIVol showed unilateral hippocampal atrophy in 60.9% of patients, bilateral asymmetric hippocampal atrophy in 19.0%, symmetric hippocampal atrophy in 3.8%, and normal volumes in 16.3%. There was a significant concordance between MRIVol lateralization and both interictal and ictal EEG lateralization (P<.001). All patients with unilateral hippocampal atrophy had concordant interictal and ictal EEG lateralization. Six (18.2%) of the 33 patients with bilateral asymmetric hippocampal atrophy had MRI lateralization discordant with EEG lateralization. CONCLUSIONS: We found a strong concordance between EEG and MRIVol lateralization in patients with TLE. Unilateral hippocampal atrophy predicted ipsilateral interictal epileptiform abnormalities and ipsilateral seizure onsets with no false lateralization. Previous studies in addition to the present series support that a concordant outpatient EEG evaluation in patients with TLE and unilateral hippocampal atrophy would obviate the need for inpatient EEG monitoring.     

Ictal scalp EEG findings in patients with mesial temporal lobe epilepsy.

The syndrome of mesial temporal lobe epilepsy (MTLE) is a well-defined clinical entity that responds to surgical treatment in a considerable number of patients. Although it has been subjected to intensive clinical research, few investigators have published the ictal scalp EEG findings and looked for specific features that might predict postoperative outcome. This study was designed to examine ictal scalp EEG characteristics in detail, in a group of patients with pathologically confirmed hippocampal sclerosis (HS). Patients who underwent long-term video-EEG monitoring at our center during a 3-year period and were diagnosed to have MTLE and pathologically proven HS were included in this retrospective study. All ictal scalp EEGs were investigated in a common reference montage, paying attention to the localization, morphology and frequency of ictal discharges that were accepted to represent a specific phase if the findings were sustained for at least 3 seconds. Any significant change in localization, morphology or frequency of discharges was said to represent a different phase. The ictal EEG patterns in different phases were later compared among seizures of different patients. In addition, the ictal EEG characteristics of the patients in Group I (Engel's classification) were compared with the ictal EEG findings in patients who were included in another group. All the patients have been followed for more than 5 years. Seventy-one ictal EEGs were investigated in 25 adult patients (11 M, 14 F). Onset patterns were lateralized in 81.7% and localized in 76% of the seizures. Thirteen different patterns of onset were detected, the most common of which was the cessation of interictal discharges (35.2%). The most common ictal pattern following the initial changes was ipsilateral temporal rhythmic theta-delta activity (85.2%) that occurred on the average 13.4 seconds after onset. Nonlocalized/lateralized seizure onset of all the seizures or bilateral independent onset was present in 75% of the patients in Groups II-III, whereas this ratio was 14.3% in the patients in Group I (p=0.031). In conclusion, ictal scalp EEG in MTLE allows correct lateralization and localization in most of the seizures. Onset patterns may vary considerably; however, a later significant pattern consisting of rhythmic ipsilateral temporal build-up develops in the majority of seizures. Some ictal EEG characteristics may be related to post-operative outcome.     

Recognition and Classification of Seizures in Infants

Summary: Purpose: We wished to assess the reliability of the International League Against Epilepsy (ILAE) seizure classification system applied to infantile seizures and to test a proposed new classification. Methods: We first analyzed 39 seizures in 20 infants (aged 1–26 months) recorded with simultaneous closed-circuit television and EEG (CCTV/EEG). EEGs and videotapes of all seizures were independently analyzed by two epileptologists blinded to clinical histories. Videotapes of each seizure were reviewed without simultaneous EEG (phase I), and printouts of ictal EEGs were assessed without behavioral correlates (phase II). The observers classified seizures according to ILAE criteria. Interrater agreement was assessed by the kappa statistic. Results: Agreement on EEG features (phase II) was moderate (= 0.54) in identifying focal ictal onsets and substantial (= 0.79) in identifying generalized onsets. In contrast, analysis of videotapes showed substantial disagreement between observers in terms of classifying seizures as partial or generalized. Therefore, agreement between observers for partial was slight (= 0.14) and fair for generalized seizures (= 0.26). Similarly, conclusions of the observers as compared with those of a consensus panel were divergent for both partial (= 0.18) and generalized seizures (= 0.30). We therefore developed an alternative classification scheme and retested interrater agreement in a review of 50 seizures in 25 other infants. With this classification scheme, there was substantial agreement between observers (= 0.72). Conclusions: With clinical observations and interictal EEGs, seizures in infants cannot be reliably classified by current ILAE criteria. In contrast, a proposed new classification scheme based solely on semiology showed substantial reliability.     

Use of cavernous sinus EEG in the detection of seizure onset and spread in mesial temporal lobe epilepsy

PURPOSE: The study goal was to evaluate the clinical usefulness of intravenous EEG recording by placing wire electrodes in the cavernous sinus (CS) and the superior petrosal sinus (SPS) in patients with intractable temporal lobe epilepsy (TLE), with special emphasis on the ictal recording. METHODS: We placed Seeker Lite-10 guide wire as electrodes in the bilateral CS, SPS, or both to simultaneously record both ictal and interictal EEGs with the scalp EEG in five patients with TLE. In addition, in one patient, we averaged interictal scalp and intravascular EEG time-locked to the epileptiform discharge recorded from the CS/SPS-EEG to further delineate the relationship of the spikes between scalp and intravenous recording. RESULTS: In four of five patients, clinically useful recording was obtained to determine ictal focus. We recorded habitual seizures in three patients, and the detailed characteristics of ictal epileptiform discharges were shown. The averaged waveform of interictal epileptiform discharges clarified the spike distribution in the scalp EEGs, which was otherwise undetectable in the single trace. All of the patients completed the intravenous EEG monitoring without any neurological or psychological problems. CONCLUSIONS: The CS/SPS-EEG is a relatively noninvasive method that is useful for the detection of ictal focus and its spreading pattern and thus for the selection of surgical candidate among patients with intractable TLE. Although the number of seizures detected during the short monitoring period may be limited, due to the advantages of its safety and simplicity, it is worth trying for potential surgical candidates before more invasive examinations are applied. A further study with a larger number of patients is needed to estimate its practical risk.     

Epileptiform patterns of children

Infants and children display almost the entire range of interictal and ictal epileptiform patterns, exclusive of those seen in the neonatal period. This paper reviews patterns seen in the idiopathic and secondary forms of epilepsies, both generalized and partial, as well as periodic patterns and benign epileptiform variants as they appear in childhood. The discussion focuses on interictal activities, the abnormalities most commonly seen in the EEG laboratory. Special attention is paid to differential diagnostic features; the distinctions between significant epileptiform patterns and normal background features of the maturing child's EEG can be particularly challenging. A large number of illustrations are provided to demonstrate not only the classic epileptiform patterns but also many of the less typical presentations, since children's EEGs show a greater number of atypical waveforms.     

Interictal, unifocal spikes in refractory extratemporal epilepsy predict ictal origin and postsurgical outcome.

OBJECTIVES: To evaluate the significance of exclusively unifocal, unilateral, interictal epileptiform patterns on scalp electroencephalography (EEG) in surgical candidates with medically intractable extratemporal epilepsy. METHODS: We reviewed 126 patients with refractory extratemporal partial seizures who underwent epilepsy surgery at our center. All were followed for at least 2 years after resections. Surgery was based on ictal EEG recordings. We examined ictal onsets and surgical outcome in subjects whose preoperative, interictal scalp EEGs during long-term monitoring (LTM) demonstrated only unilateral, well-defined focal discharges, and outcome in patients whose interictal EEGs during LTM showed bilateral, non-localized, or multifocal epileptiform patterns. RESULTS: We found that 26 subjects exhibited only unilateral, unifocal, interictal epileptiform patterns. In all 26 cases (100%) clinical seizures arose from the regions expected by the interictal findings (P<0.0001, Sign test). At last follow-up 77% (20/26) of these patients were seizure-free, while 23% (6/26) had >75% reduction in seizures. This compares to the remaining patients, of whom 34% (34/100) were seizure-free, 41% (41/100) had >75% reduction in seizures, and 25% (25/100) had <75% reduction in seizures (P=0.0001, Fisher's Exact test). CONCLUSIONS: Strictly unifocal, interictal epileptiform patterns on scalp EEG, though seen in a minority of subjects, may be an important, independent factor in evaluating subjects with intractable extratemporal, localization-related epilepsy for surgical therapy. This finding is highly predictive of both ictal onsets and successful postsurgical outcome.     

Seizure prediction with spectral power of EEG using cost-sensitive support vector machines

Purpose: We propose a patient‐specific algorithm for seizure prediction using multiple features of spectral power from electroencephalogram (EEG) and support vector machine (SVM) classification. Methods: The proposed patient‐specific algorithm consists of preprocessing, feature extraction, SVM classification, and postprocessing. Preprocessing removes artifacts of intracranial EEG recordings and they are further preprocessed in bipolar and/or time‐differential methods. Features of spectral power of raw, or bipolar and/or time‐differential intracranial EEG (iEEG) recordings in nine bands are extracted from a sliding 20‐s–long and half‐overlapped window. Nine bands are selected based on standard EEG frequency bands, but the wide gamma bands are split into four. Cost‐sensitive SVMs are used for classification of preictal and interictal samples, and double cross‐validation is used to achieve in‐sample optimization and out‐of‐sample testing. We postprocess SVM classification outputs using the Kalman Filter and it removes sporadic and isolated false alarms. The algorithm has been tested on iEEG of 18 patients of 20 available in the Freiburg EEG database who had three or more seizure events. To investigate the discriminability of the features between preictal and interictal, we use the Kernel Fisher Discriminant analysis. Key findings: The proposed patient‐specific algorithm for seizure prediction has achieved high sensitivity of 97.5% with total 80 seizure events and a low false alarm rate of 0.27 per hour and total false prediction times of 13.0% over a total of 433.2 interictal hours by bipolar preprocessing (92.5% sensitivity, a false positive rate of 0.20 per hour, and false prediction times of 9.5% by time‐differential preprocessing). This high prediction rate demonstrates that seizures can be predicted by the patient‐specific approach using linear features of spectral power and nonlinear classifiers. Bipolar and/or time‐differential preprocessing significantly improves sensitivity and specificity. Spectral powers in high gamma bands are the most discriminating features between preictal and interictal. Significance: High sensitivity and specificity are achieved by nonlinear classification of linear features of spectral power. Power changes in certain frequency bands already demonstrated their possibilities for seizure prediction indicators, but we have demonstrated that combining those spectral power features and classifying them in a multivariate approach led to much higher prediction rates. Employing only linear features is advantageous, especially when it comes to an implantable device, because they can be computed rapidly with low power consumption.     

Preoperative EEG correlates of seizure outcome from epilepsy surgery in children

To identify criteria for selecting pediatric candidates for cortical resection, we compared preoperative standard EEGs, extracranial or intracranial (subdural strip or grid) electrode EEG-video monitoring and imaging studies, and surgical pathology findings with long-term (mean = 6.1 years) seizure outcome. Of 47 children, 83% were either seizure-free or substantially improved. Children with a predominant single interictal focus in standard EEGs over time or with seizures that were mostly well localized (even with equal numbers of bilaterally independent spikes) during monitoring had significantly better outcomes than those who had mostly diffuse or bilaterally independent interictal and ictal discharges. Factors not associated with poor outcomes were some seizures unaccompanied by ictal EEG patterns, spikes in postresection electrocorticograms, extratemporal resection, younger age at surgery, and older age of seizure onset. Significantly more patients with than without tumors were seizure-free. Significantly more preteenagers than teenagers had poorly localized interictal and ictal discharges.     

Improving the interpretation of ictal scalp EEG: BSS-CCA algorithm for muscle artifact removal

PURPOSE: To investigate the potential clinical relevance of a new algorithm to remove muscle artifacts in ictal scalp EEG. METHODS: Thirty-seven patients with refractory partial epilepsy with a well-defined seizure onset zone based on full presurgical evaluation, including SISCOM but excluding ictal EEG findings, were included. One ictal EEG of each patient was presented to a clinical neurophysiologist who was blinded to all other data. Ictal EEGs were first rated after band-pass filtering, then after elimination of muscle artifacts using a blind source separation-canonical correlation analysis technique (BSS-CCA). Degree of muscle artifact contamination, lateralization, localization, time and pattern of ictal EEG onset were compared between the two readings and validated against the other localizing information. RESULTS: Muscle artifacts contaminated 97% of ictal EEGs, and interfered with the interpretation in 76%, more often in extratemporal than temporal lobe seizures. BSS-CCA significantly improved the sensitivity to localize the seizure onset from 62% to 81%, and performed best in ictal EEGs with moderate to severe muscle artifact contamination. In a significant number of the contaminated EEGs, BSS-CCA also led to an earlier identification of ictal EEG changes, and recognition of ictal EEG patterns that were hidden by muscle artifact. CONCLUSIONS: Muscle artifacts interfered with the interpretation in a majority of ictal EEGs. BSS-CCA reliably removed these muscle artifacts in a user-friendly manner. BSS-CCA may have an important place in the interpretation of ictal EEGs during presurgical evaluation of patients with refractory partial epilepsy.     

Independent component analysis removing artifacts in ictal recordings

PURPOSE: Independent component analysis (ICA) is a novel algorithm able to separate independent components from complex signals. Studies in interictal EEG demonstrate its usefulness to eliminate eye, muscle, 50-Hz, electrocardiogram (ECG), and electrode artifacts. The goal of this study was to evaluate the usefulness of ICA in removing artifacts in ictal recordings with a known EEG onset. METHODS: We studied 20 seizures of nine patients with focal epilepsy monitored in our video-EEG monitoring unit. ICA was applied to remove obvious artifacts in segments at the beginning of the seizure. The final EEGs were exported to the original format and were compared with the original EEG by two blinded examiners. We compared original recordings and the samples cleaned by digital filters (DFs), ICA and ICA plus digital filters (ICA + DFs), evaluating the possibility of finding an ictal pattern, the localization of the onset in area and time, and the global quality of the sample. RESULTS: All the recordings except one (95%) improved after the use of ICA for the elimination of blinking and other artifacts. Three seizures were found in which in the original recordings did not permit us to detect an ictal pattern, and after ICA + DFs, an ictal onset was evident; in two of them, ICA alone was able to show this pattern. The best results in all the scores were obtained with ICA + DF. ICA was better than DFs. The agreement between the two reviewers was highly significant. CONCLUSIONS: ICA is useful to remove artifacts from ictal recordings. When applied to ictal recordings, it increases the quality of the recording. In some cases, ICA may be useful to show ictal onsets obscured by artifacts. ICA + DFs obtained the best results regarding removal of the artifacts.     

Significance of Simple Partial Seizures in Temporal Lobe Epilepsy

Summary: We determined how localization of simple partial seizures (SPS) correlated with localization of complex partial seizure (CPS) in scalp/sphenoidal EEG and assessed prognosis after temporal lobe resective surgery in patients with an ictal correlate of SPS in scalp/sphenoidal EEG recordings. EEGs were recorded with the 10–20 system of electrode placement and supplemented with sphenoidal electrodes. Between 1985 and 1992, 183 patients with temporal lobe epilepsy (TLE) reported an aura (SPS) during inpatient monitoring; all were eligible for inclusion in our study. The EEGs during SPS showed ictal changes in 51 patients (28%, 117 SPS). Forty-four patients had unilateral temporal interictal spikes (IIS), and SPS and CPS always arose from the same region. Seven patients had bitemporal interictal spikes; SPS colocalized with CPS in 4 patients (57%), SPS were contralateral to CPS in 2 patients, and 1 patient had bilateral independent CPS but unilateral SPS. SPS accompanied by EEG ictal changes conveyed a favorable prognosis in patients who underwent epilepsy surgery. Scalp/sphenoidal recorded IIS but were less reliable in identifying the location of CPS onset in patients with bitemporal spikes.     

New diagnostic EEG markers of the Alzheimer’s disease using visibility graph

A new chaos–wavelet approach is presented for electroencephalogram (EEG)-based diagnosis of Alzheimer’s disease (AD) employing a recently developed concept in graph theory, visibility graph (VG). The approach is based on the research ideology that nonlinear features may not reveal differences between AD and control group in the band-limited EEG, but may represent noticeable differences in certain sub-bands. Hence, complexity of EEGs is computed using the VGs of EEGs and EEG sub-bands produced by wavelet decomposition. Two methods are employed for computation of complexity of the VGs: one based on the power of scale-freeness of a graph structure and the other based on the maximum eigenvalue of the adjacency matrix of a graph. Analysis of variation is used for feature selection. Two classifiers are applied to the selected features to distinguish AD and control EEGs: a Radial Basis Function Neural Network (RBFNN) and a two-stage classifier consisting of Principal Component Analysis (PCA) and the RBFNN. After comprehensive statistical studies, effective classification features and mathematical markers were discovered. Finally, using the discovered features and a two-stage classifier (PCA-RBFNN), a high diagnostic accuracy of 97.7% was obtained.     

Ictal and interictal EEG patterns in patients with nonconvulsive and subtle convulsive status epilepticus

BackgroundElectroencephalography findings in nonconvulsive or subtle convulsive status epilepticus (NCSE and SCSE, respectively) can be heterogenous. We aimed to study the different patterns on EEG in our cohort of patients.ObjectiveOur objective was to study ictal and interictal EEG patterns in patients with NCSE and SCSE.MethodsFrom January 2012 to December 2013, EEGs recorded from patients admitted for altered mental status suspected of having NCSE or SCSE were reviewed retrospectively. Electroencephalography status was defined as having (a) continuous ictal discharges lasting > 5 min or (b) > 2 discrete bursts of ictal discharges, each lasting < 5 min, without returning to previous background rhythm in between these bursts.ResultsAmong 1698 EEGs recorded for at least 30 min from hospitalized patients, 55 (3.23%) satisfied the criteria of EEG SE. The ictal onset was regional in 37 (67.2%) EEGs, multiregional independent in 8 (14.5%), and generalized in 10 (18.4%).The EEG seizure duration was > 5 min in 24 (43.6%) EEGs, between 1 and 5 min in 14 (25.4%), and less than 1 min in 17 (30.8%).Twenty (36.3%) EEGs showed one continuous prolonged seizure episode of > 5-minute duration, 15 (27.2%) had 10 or less discrete episodes, 20 (36.3%) had more than 10 episodes, and 11 (20%) had 2 or more ictal patterns.Thirty (54.5%) EEGs had onset ictal frequency of > 8 Hz whereas the rest had < 8-Hz ictal frequency. In the interictal segment, 29 patients had continuous generalized slow waves, while 12 had intermittent generalized slow waves. Eleven patients had continuous slow waves lateralized to one hemisphere, and these were ipsilateral to the ictal focus in 10 but contralateral in 1. Other interictal waves seen were PLEDS (6), sharp waves (3), suppression (5), and triphasic waves (1).The background alpha rhythm was absent in 36 patients and slow in 14, and normal background alpha was seen in the interictal period in 5 patients.ConclusionThe ictal and interictal EEG patterns in NCSE and SCSE can be varied. Further study to look for etiologic and clinical correlates of each pattern could add to its clinical value.This article is part of a Special Issue entitled “Status Epilepticus”.