Nathan J. Stevenson

A Nonstationary Model of Newborn EEG

The detection of seizure in the newborn is a critical aspect of neurological research. Current automatic detection techniques are difficult to assess due to the problems associated with acquiring and labelling newborn electroencephalogram (EEG) data. A realistic model for newborn EEG would allow confident development, assessment and comparison of these detection techniques. This paper presents a model for newborn EEG that accounts for its self-similar and nonstationary nature. The model consists of background and seizure submodels. The newborn EEG background model is based on the short-time power spectrum with a time-varying power law. The relationship between the fractal dimension and the power law of a power spectrum is utilized for accurate estimation of the short-time power law exponent. The newborn EEG seizure model is based on a well-known time-frequency signal model. This model addresses all significant time-frequency characteristics of newborn EEG seizure which include; multiple components or harmonics, piecewise linear instantaneous frequency laws and harmonic amplitude modulation. Estimates of the parameters of both models are shown to be random and are modelled using the data from a total of 500 background epochs and 204 seizure epochs. The newborn EEG background and seizure models are validated against real newborn EEG data using the correlation coefficient. The results show that the output of the proposed models have a higher correlation with real newborn EEG than currently accepted models (a 10% and 38% improvement for background and seizure models, respectively)

Cooling and seizure burden in term neonates: an observational study.

Objective To investigate any possible effect of cooling on seizure burden, the authors quantified the recorded electrographic seizure burden based on multichannel video-EEG recordings in term neonates with hypoxic-ischaemic encephalopathy (HIE) who received cooling and in those who did not. Study design Retrospective observational study. Patients Neonates >37 weeks gestation born between 2003 and 2010 in two hospitals. Methods Off-line analysis of prolonged continuous multichannel video-EEG recordings was performed independently by two experienced encephalographers. Comparison between the recorded electrographic seizure burden in non-cooled and cooled neonates was assessed. Data were treated as non-parametric and expressed as medians with interquartile ranges (IQR). Results One hundred and seven neonates with HIE underwent prolonged continuous multichannel EEG monitoring. Thirty-seven neonates had electrographic seizures, of whom 31 had EEG recordings that were suitable for the analysis (16 non-cooled and 15 cooled). Compared with non-cooled neonates, multichannel EEG monitoring commenced at an earlier postnatal age in cooled neonates (6 (3–9) vs 15 (5–20) h)and continued for longer (88 (75–101) vs 55 (41–60) h). Despite this increased opportunity to capture seizures in cooled neonates, the recorded electrographic seizure burden in the cooled group was significantly lower than in the non-cooled group (60 (39–224) vs 203 (141–406) min). Further exploratory analysis showed that the recorded electrographic seizure burden was only significantly reduced in cooled neonates with moderate HIE (49 (26–89) vs 162 (97–262) min). Conclusions A decreased seizure burden was seen in neonates with moderate HIE who received cooling. This finding may explain some of the therapeutic benefits of cooling seen in term neonates with moderate HIE.

The temporal evolution of electrographic seizure burden in neonatal hypoxic ischemic encephalopathy

Purpose:  Hypoxic ischemic encephalopathy (HIE) accounts for 60% of all neonatal seizures. There is emerging evidence that seizures cause additional injury to the developing brain that has sustained hypoxic ischemic injury. Temporal evolution of clinical seizure burden in HIE has been characterized, with maximum clinical seizure burden (the period of maximum seizure activity) being observed between 12 and 24 h of age. The purpose of our study was to investigate the distribution of electrographic seizure burden (the accumulated duration of seizures over a defined time period), following the initial hypoxic ischemic insult.

A nonparametric feature for neonatal EEG seizure detection based on a representation of pseudo-periodicity ✩

Automated methods of neonatal EEG seizure detection attempt to highlight the evolving, stereotypical, pseudo-periodic, nature of EEG seizure while rejecting the nonstationary, modulated, coloured stochastic background in the presence of various EEG artefacts. An important aspect of neonatal seizure detection is, therefore, the accurate representation and detection of pseudo-periodicity in the neonatal EEG. This paper describes a method of detecting pseudo-periodic components associated with neonatal EEG seizure based on a novel signal representation; the nonstationary frequency marginal (NFM). The NFM can be considered as an alternative time-frequency distribution (TFD) frequency marginal. This method integrates the TFD along data-dependent, time-frequency paths that are automatically extracted from the TFD using an edge linking procedure and has the advantage of reducing the dimension of a TFD. The reduction in dimension simplifies the process of estimating a decision statistic designed for the detection of the pseudo-periodicity associated with neonatal EEG seizure. The use of the NFM resulted in a significant detection improvement compared to existing stationary and nonstationary methods. The decision statistic estimated using the NFM was then combined with a measurement of EEG amplitude and nominal pre- and post-processing stages to form a seizure detection algorithm. This algorithm was tested on a neonatal EEG database of 18 neonates, 826 h in length with 1389 seizures, and achieved comparable performance to existing second generation algorithms (a median receiver operating characteristic area of 0.902; IQR 0.835-0.943 across 18 neonates).

Quantitative EEG analysis in neonatal hypoxic ischaemic encephalopathy.

OBJECTIVE To test the hypothesis that quantitative EEG (qEEG) measures are associated with a grading of HIE based on the visual interpretation of neonatal EEG (EEG/HIE). METHODS Continuous multichannel video-EEG data were recorded for up to 72 h. One-hour EEG segments from each recording were visually analysed and graded by two electroencephalographers (EEGers) blinded to clinical data. Several qEEG measures were calculated for each EEG segment. Kruskal-Wallis testing with post hoc analysis and multiple linear regression were used to assess the hypothesis. RESULTS Fifty-four full-term infants with HIE were studied. The relative delta power, skewness, kurtosis, amplitude, and discontinuity were significantly different across four EEG/HIE grades (p<0.05). A linear combination of these qEEG measures could predict the EEG/HIE grade assigned by the EEGers with an accuracy of 89%. CONCLUSION Quantitative analysis of background EEG activity has shown that measures based on the amplitude, frequency content and continuity of the EEG are associated with a visual interpretation of the EEG performed by experienced EEGers. SIGNIFICANCE Identifying qEEG measures that can separate between EEG/HIE grades is an important first step towards creating a classifier for automated detection of EEG/HIE grades.

A Nonlinear Model of Newborn EEG with Nonstationary Inputs

Newborn EEG is a complex multiple channel signal that displays nonstationary and nonlinear characteristics. Recent studies have focussed on characterizing the manifestation of seizure on the EEG for the purpose of automated seizure detection. This paper describes a novel model of newborn EEG that can be used to improve seizure detection algorithms. The new model is based on a nonlinear dynamic system; the Duffing oscillator. The Duffing oscillator is driven by a nonstationary impulse train to simulate newborn EEG seizure and white Gaussian noise to simulate newborn EEG background. The use of a nonlinear dynamic system reduces the number of parameters required in the model and produces more realistic, life-like EEG compared with existing models. This model was shown to account for 54% of the linear variation in the time domain, for seizure, and 85% of the linear variation in the frequency domain, for background. This constitutes an improvement in combined performance of 6%, with a reduction from 48 to 4 model parameters, compared to an optimized implementation of the best performing existing model.

Seizure burden and neurodevelopmental outcome in neonates with hypoxic-ischemic encephalopathy.

To examine the relationship between electrographic seizures and long‐term outcome in neonates with hypoxic–ischemic encephalopathy (HIE).

Early postnatal EEG features of perinatal arterial ischaemic stroke with seizures.

Background Stroke is the second most common cause of seizures in term neonates and is associated with abnormal long-term neurodevelopmental outcome in some cases. Objective To aid diagnosis earlier in the postnatal period, our aim was to describe the characteristic EEG patterns in term neonates with perinatal arterial ischaemic stroke (PAIS) seizures. Design Retrospective observational study. Patients Neonates >37 weeks born between 2003 and 2011 in two hospitals. Method Continuous multichannel video-EEG was used to analyze the background patterns and characteristics of seizures. Each EEG was assessed for continuity, symmetry, characteristic features and sleep cycling; morphology of electrographic seizures was also examined. Each seizure was categorized as electrographic-only or electroclinical; the percentage of seizure events for each seizure type was also summarized. Results Nine neonates with PAIS seizures and EEG monitoring were identified. While EEG continuity was present in all cases, the background pattern showed suppression over the infarcted side; this was quite marked (>50% amplitude reduction) when the lesion was large. Characteristic unilateral bursts of theta activity with sharp or spike waves intermixed were seen in all cases. Sleep cycling was generally present but was more disturbed over the infarcted side. Seizures demonstrated a characteristic pattern; focal sharp waves/spike-polyspikes were seen at frequency of 1–2 Hz and phase reversal over the central region was common. Electrographic-only seizure events were more frequent compared to electroclinical seizure events (78 vs 22%). Conclusions Focal electrographic and electroclinical seizures with ipsilateral suppression of the background activity and focal sharp waves are strong indicators of PAIS. Approximately 80% of seizure events were the result of clinically unsuspected seizures in neonates with PAIS. Prolonged and continuous multichannel video-EEG monitoring is advocated for adequate seizure surveillance.

Validation of an automated seizure detection algorithm for term neonates.

Highlights • Seizure detection algorithm (SDA) validated on unseen, unedited EEG of 70 neonates.• Results at SDA sensitivity settings of 0.5–0.3 acceptable for clinical use.• Seizure detection rate of 52.6–75.0%, false detection rate 0.04–0.36 FD/h.

Interobserver agreement for neonatal seizure detection using multichannel EEG.

To determine the interobserver agreement (IOA) of neonatal seizure detection using the gold standard of conventional, multichannel EEG.