Improved epileptic seizure detection combining dynamic feature normalization with EEG novelty detection.

Continuous electroencephalographic monitoring of critically ill patients is an established procedure in intensive care units. Seizure detection algorithms, such as support vector machines (SVM), play a prominent role in this procedure. To correct for inter-human differences in EEG characteristics, as well as for intra-human EEG variability over time, dynamic EEG feature normalization is essential. Recently, the median decaying memory (MDM) approach was determined to be the best method of normalization. MDM uses a sliding baseline buffer of EEG epochs to calculate feature normalization constants. However, while this method does include non-seizure EEG epochs, it also includes EEG activity that can have a detrimental effect on the normalization and subsequent seizure detection performance. In this study, EEG data that is to be incorporated into the baseline buffer are automatically selected based on a novelty detection algorithm (Novelty-MDM). Performance of an SVM-based seizure detection framework is evaluated in 17 long-term ICU registrations using the area under the sensitivity-specificity ROC curve. This evaluation compares three different EEG normalization methods, namely a fixed baseline buffer (FB), the median decaying memory (MDM) approach, and our novelty median decaying memory (Novelty-MDM) method. It is demonstrated that MDM did not improve overall performance compared to FB (p < 0.27), partly because seizure like episodes were included in the baseline. More importantly, Novelty-MDM significantly outperforms both FB (p = 0.015) and MDM (p = 0.0065).

Optimal training dataset composition for SVM-based, age-independent, automated epileptic seizure detection.

Automated seizure detection is a valuable asset to health professionals, which makes adequate treatment possible in order to minimize brain damage. Most research focuses on two separate aspects of automated seizure detection: EEG feature computation and classification methods. Little research has been published regarding optimal training dataset composition for patient-independent seizure detection. This paper evaluates the performance of classifiers trained on different datasets in order to determine the optimal dataset for use in classifier training for automated, age-independent, seizure detection. Three datasets are used to train a support vector machine (SVM) classifier: (1) EEG from neonatal patients, (2) EEG from adult patients and (3) EEG from both neonates and adults. To correct for baseline EEG feature differences among patients feature, normalization is essential. Usually dedicated detection systems are developed for either neonatal or adult patients. Normalization might allow for the development of a single seizure detection system for patients irrespective of their age. Two classifier versions are trained on all three datasets: one with feature normalization and one without. This gives us six different classifiers to evaluate using both the neonatal and adults test sets. As a performance measure, the area under the receiver operating characteristics curve (AUC) is used. With application of FBC, it resulted in performance values of 0.90 and 0.93 for neonatal and adult seizure detection, respectively. For neonatal seizure detection, the classifier trained on EEG from adult patients performed significantly worse compared to both the classifier trained on EEG data from neonatal patients and the classier trained on both neonatal and adult EEG data. For adult seizure detection, optimal performance was achieved by either the classifier trained on adult EEG data or the classifier trained on both neonatal and adult EEG data. Our results show that age-independent seizure detection is possible by training one classifier on EEG data from both neonatal and adult patients. Furthermore, our results indicate that for accurate age-independent seizure detection, it is important that EEG data from each age category are used for classifier training. This is particularly important for neonatal seizure detection. Our results underline the under-appreciated importance of training dataset composition with respect to accurate age-independent seizure detection.

EEG feature pre-processing for neonatal epileptic seizure detection.

Aim of our project is to further optimize neonatal seizure detection using support vector machine (SVM). First, a Kalman filter (KF) was used to filter both feature and classifier output time series in order to increase temporal precision. Second, EEG baseline feature correction (FBC) was introduced to reduce inter patient variability in feature distributions. The performance of the detection methods is evaluated on 54 multi channel routine EEG recordings from 39 both term and pre-term newborns. The area under the receiver operating characteristics curve (AUC) as well as sensitivity and specificity are used to evaluate the performance of the classification method. SVM without KF and FBC achieves an AUC of 0.767 (sensitivity 0.679, specificity 0.707). The highest AUC of 0.902 (sensitivity 0.801, specificity 0.831) is achieved on baseline corrected features with a Kalman smoother used for training data pre-processing and a KF used to filter the classifier output. Both FBC and KF significantly improve neonatal epileptic seizure detection. This paper introduces significant improvements for the state of the art SVM based neonatal epileptic seizure detection.

[Anti-NMDA-receptor encephalitis: a new axis-III disorder in the differential diagnosis of childhood disintegrative disorder, early onset schizophrenia and late onset autism]. / Anti-NMDA-receptorencefalitis: een nieuwe as III-stoornis in de differentiaaldiagnose van desintegratiestoornis van de kinderleeftijd, vroeg optredende schizofrenie en laat optredend autisme.

Childhood disintegrative disorder (CDD), early onset schizophrenia (EOS), and late onset autism (LOA) often follow a similar course: initially, development is normal, then there is a sudden neuropsychiatric deterioration of social interaction and communication skills, which is combined with a decline in intelligence and reduction in daily activities. A 9-year-old boy was admitted to the paediatric ward with acute onset of secondary epileptic seizures. It was not long until the boy's symptoms resembled that of patients with cdd, eos and loa. Intensive tests led to the diagnosis of anti-NMDA-receptor encephalitis. Anti-NMDA-receptor encephalitis should be regarded as a possible organic cause underlying the syndromal presentation of CDD, EOS and LOA.

40-Hz steady state response in Alzheimer's disease and mild cognitive impairment.

The 40-Hz steady state response (SSR) reflects early sensory processing and can be measured with electroencephalography (EEG). The current study compared the 40-Hz SSR in groups consisting of mild Alzheimer's disease patients (AD) (n=15), subjects with mild cognitive impairment (MCI) (n=20) and healthy elderly control subjects (n=20). All participants were naïve for psychoactive drugs. Auditory click trains at a frequency of 40-Hz evoked the 40-Hz SSR. To evaluate test-retest reliability (TRR), subjects underwent a similar assessment 1 week after the first. The results showed a high TRR and a significant increase of 40-Hz SSR power in the AD group compared to MCI and controls. Furthermore a moderate correlation between 40-Hz SSR power and cognitive performance as measured by ADAS-cog was shown. The results suggest that 40-Hz SSR might be an interesting candidate marker of disease progression.

Increased EEG gamma band activity in Alzheimer's disease and mild cognitive impairment.

High frequency (30-70 Hz) gamma band oscillations in the human electro-encephalogram (EEG) are thought to reflect perceptual and cognitive processes. It is therefore interesting to study these measures in cognitive impairment and dementia. To evaluate gamma band oscillations as a diagnostic biomarker in Alzheimer's disease (AD) and mild cognitive impairment (MCI), 15 psychoactive drug naïve AD patients, 20 MCI patients and 20 healthy controls participated in this study. Gamma band power (GBP) was measured in four conditions viz. resting state, music listening, story listening and visual stimulation. To evaluate test-retest reliability (TRR), subjects underwent a similar assessment one week after the first. The overall TRR was high. Elevated GBP was observed in AD when compared to MCI and control subjects in all conditions. The results suggest that elevated GBP is a reproducible and sensitive measure for cognitive dysfunction in AD in comparison with MCI and controls.

The ictal bradycardia syndrome.

The ictal bradycardia syndrome is an uncommon diagnosis in which bradycardia is accompanied by simultaneous epileptic discharges in the EEG. We describe a patient who was referred to the emergency ward because of syncope. Ictal semeiology and EEG-EG findings are discussed and compared with those published in the literature. Therapeutic options are discussed in relation with those published in the literature. The ictal bradycardia syndrome is probably underdiagnosed, while its recognition is of utmost importance because of potential life threatening complications such as asystole. Up to now, its aetiology is poorly understood, its ictal semeiology is often described insufficiently and its therapy is still discussed.

Placebo controlled utility and feasibility study of the H-reflex and flexor reflex in spastic children treated with intrathecal baclofen.

OBJECTIVE: To evaluate feasibility and utility of the soleus H-reflex and tibialis anterior flexor reflex (FR) in identifying spinal cord neuronal response to intrathecal baclofen (ITB) in children with severe spastic cerebral palsy. METHODS: During a randomized, double-blind, placebo-controlled dose-escalation test treatment, maximum H amplitude/maximum M amplitude (H/M ratio) and FR parameters were bilaterally recorded at baseline and 2-3 h after intrathecal bolus administration of placebo and increasing doses of baclofen until both an improvement in the individual treatment goal(s) and a one-point reduction on the Ashworth scale were observed. RESULTS: Electrophysiological data of 14 children were studied. The H-reflex was feasible in 13 children, the FR threshold area in 9 and the FR, elicited with supramaximal stimulation, in only one child. After ITB, the H/M ratio significantly decreased (left: 0.67+/-0.47 to 0.15+/-0.18, P=0.005; right: 0.55+/-0.32 to 0.14+/-0.19, P=0.002) without placebo effect. FR threshold area after ITB, only decreased significantly in children not taking oral baclofen (left: 146+/-53 to 41+/-54 mV ms, P=0.000; right: 156+/-80 to 66+/-48 mV ms, P=0.002). CONCLUSIONS: This is the first randomized, double-blind, placebo-controlled dose-escalation study in spastic children demonstrating the soleus H-reflex to be a feasible and objective measure to quantify the decreasing motoneuron excitability in response to ITB bolus administration. Only in children not taking oral baclofen, FR threshold area can also be used as an objective outcome measure, yet feasibility is limited. SIGNIFICANCE: We suggest introducing the H-reflex as the electrophysiological gold standard for the evaluation of the effect of ITB in spastic children.