A seizure prediction algorithm is proposed that combines novel multivariate EEG features with patient-specific machine learning. The algorithm computes the eigenspectra of space-delay correlation and covariance matrices from 15-s blocks of EEG data at multiple delay scales. The principal components of these features are used to classify the patient's preictal or interictal state. This is done using a support vector machine (SVM), whose outputs are averaged using a running 15-minute window to obtain a final prediction score. The algorithm was tested on 19 of 21 patients in the Freiburg EEG data set who had three or more seizures, predicting 71 of 83 seizures, with 15 false predictions and 13.8 h in seizure warning during 448.3 h of interictal data. The proposed algorithm scales with the number of available EEG signals by discovering the variations in correlation structure among any given set of signals that correlate with seizure risk.

Summary:Purpose: To study the short-term effects of vagus nerve stimulation (VNS) on brain activation and cerebral blood flow by using functional magnetic resonance imaging (fMRI).Methods: Five patients (three women, two men; mean age, 35.4 years) who were treated for medically refractory epilepsy with VNS, underwent fMRI. All patients had a nonfocal brain MRI. The VNS was set at 30 Hz, 0.5-2.0 mA for intervals of activation of 30 s on and 30 s off, during which the fMRI was performed. Statistical parametric mapping (SPM) was used to determine significant areas of activation or inhibition during vagal nerve stimulation (p < 0.05).Results: VNS-induced activation was detected in the thalami bilaterally (left more than right), insular cortices bilaterally, ipsilateral basal ganglia and postcentral gyri, right posterior superior temporal gyrus, and inferomedial occipital gyri (left more than right). The most robust activation was seen in the thalami (left more than right) and insular cortices. Conclusions: VNS-induced thalamic and insular cortical activation during fMRI suggests that these areas may play a role in modulating cerebral cortical activity, and the observed decrease in seizure frequency in patients who are given VNS may be a consequence of this increased activation.

Summary: Purpose: Nonconvulsive status epilepticus (NCSE)is an under-recognized cause of altered mental status. There are hardly any reported data on NCSE in developing countries.Material and Methods: Prospectively 210 consecutive patients with altered mental status admitted to neurological intensive care unit (NICU) of a tertiary care center in south India were studied for the frequency of NCSE. All patients were evaluated initially with 60-min emergent EEG (EmEEG) and subsequently by continuous EEG (cEEG) monitoring.Results: Of the 210 with altered mental status admitted to NICU, the diagnosis of NCSE was established in 22 (10.5%) patients, in 12 (55%) patients with 60-min EmEEG and in 10 (45%) after cEEG monitoring for 12 to 48 hours.Of the 22 patients with NCSE, 32% had subtle motor phenomena, these were not an initial presenting features, but were apparent during cEEG recording. Acute medical or neurologic etiology was the risk factor in 68% of patients. Central nervous system (CNS) infections and cortical sino-venous thrombosis (CSVT), respectively, accounted for 23% and 14% of the etiologies. Intravenous midazolam terminated NCSE in 19 patients and valproate in 2. Of the 15 patients with acute symptomatic NCSE, 4 (18%) had poor prognosis (3 deaths and one persistent vegetative state). The etiological risk factors in the 9 (41%) patients with excellent outcome included epilepsy (3), remote symptomatic (2), cryptogenic (1), and metabolic and drugs (3).Conclusions: The frequency of NCSE in the current study was comparable with those in prior reports from developed countries. CNS infections accounted for about a fifth of the etiology. Outcome was excellent in patients with nonacute symptomatic NCSE. Initial 60-min EmEEG may be performed in establishing the diagnosis of NCSE, but almost half of patients with NCSE will be missed with this approach.

The study suggests the need to consider a change in EEG strategy to assess interictal epileptiform activity. The greatest probability of capturing an interictal abnormality within 20min was in individuals with generalized epilepsy. In individuals with suspected epilepsy in whom electrographic interictal spike confirmation is deemed necessary, after a first nonspecific or normal routine EEG, a 24h EEG should be the next step in the electrographic assessment. This study suggests that there may not be much benefit in monitoring for durations longer than 24h, unless capturing a seizure is the intent.