Automatic multimodal detection for long-term seizure documentation in epilepsy.

OBJECTIVE: This study investigated sensitivity and false detection rate of a multimodal automatic seizure detection algorithm and the applicability to reduced electrode montages for long-term seizure documentation in epilepsy patients. METHODS: An automatic seizure detection algorithm based on EEG, EMG, and ECG signals was developed. EEG/ECG recordings of 92 patients from two epilepsy monitoring units including 494 seizures were used to assess detection performance. EMG data were extracted by bandpass filtering of EEG signals. Sensitivity and false detection rate were evaluated for each signal modality and for reduced electrode montages. RESULTS: All focal seizures evolving to bilateral tonic-clonic (BTCS, n=50) and 89% of focal seizures (FS, n=139) were detected. Average sensitivity in temporal lobe epilepsy (TLE) patients was 94% and 74% in extratemporal lobe epilepsy (XTLE) patients. Overall detection sensitivity was 86%. Average false detection rate was 12.8 false detections in 24h (FD/24h) for TLE and 22 FD/24h in XTLE patients. Utilization of 8 frontal and temporal electrodes reduced average sensitivity from 86% to 81%. CONCLUSION: Our automatic multimodal seizure detection algorithm shows high sensitivity with full and reduced electrode montages. SIGNIFICANCE: Evaluation of different signal modalities and electrode montages paces the way for semi-automatic seizure documentation systems.

Reduced electrode arrays for the automated detection of rhythmic and periodic patterns in the intensive care unit: Frequently tried, frequently failed?

OBJECTIVE: To investigate the effect of systematic electrode reduction from a common 10-20 EEG system on pattern detection sensitivity (SEN). METHODS: Two reviewers rated 17130 one-minute segments of 83 prospectively recorded cEEGs according to the ACNS standardized critical care EEG terminology (CCET), including burst suppression patterns (BS) and unequivocal electrographic seizures. Consensus annotations between reviewers were used as a gold standard to determine pattern detection SEN and specificity (SPE) of a computational algorithm (baseline, 19 electrodes). Electrodes were than reduced one by one in four different variations. SENs and SPEs were calculated to determine the most beneficial assembly with respect to the number and location of electrodes. RESULTS: High automated baseline SENs (84.99-93.39%) and SPEs (90.05-95.6%) were achieved for all patterns. Best overall results in detecting BS and CCET patterns were found using the "hairline+vertex" montage. While the "forehead+behind ear" montage showed an advantage in detecting ictal patterns, reaching a 15% drop of SEN with 10 electrodes, all montages could detect BS sufficiently if at least nine electrodes were available. CONCLUSION: For the first time an automated approach was used to systematically evaluate the effect of electrode reduction on pattern detection SEN in cEEG. SIGNIFICANCE: Prediction of the expected detection SEN of specific EEG patterns with reduced EEG montages in ICU patients.

Applicability of NeuroTrend as a bedside monitor in the neuro ICU.

OBJECTIVE: To assess whether ICU caregivers can correctly read and interpret continuous EEG (cEEG) data displayed with the computer algorithm NeuroTrend (NT) with the main attention on seizure detection and determination of sedation depth. METHODS: 120 screenshots of NT (480h of cEEG) were rated by 18 briefly trained nurses and biomedical analysts. Multirater agreements (MRA) as well as interrater agreements (IRA) compared to an expert opinion (EXO) were calculated for items such as pattern type, pattern location, interruption of recording, seizure suspicion, consistency of frequency, seizure tendency and level of sedation. RESULTS: MRA as well as IRA were almost perfect (80-100%) for interruption of recording, spike-and-waves, rhythmic delta activity and burst suppression. A substantial agreement (60-80%) was found for electrographic seizure patterns, periodic discharges and seizure suspicion. Except for pattern localization (70.83-92.26%), items requiring a precondition and especially those who needed interpretation like consistency of frequency (47.47-79.15%) or level of sedation (41.10%) showed lower agreements. CONCLUSIONS: The present study demonstrates that NT might be a useful bedside monitor in cases of subclinical seizures. Determination of correct sedation depth by ICU caregivers requires a more detailed training. SIGNIFICANCE: Computer algorithms may reduce the workload of cEEG analysis in ICU patients.

Automatic detection of rhythmic and periodic patterns in critical care EEG based on American Clinical Neurophysiology Society (ACNS) standardized terminology.

AIMS OF THE STUDY: Continuous EEG from critical care patients needs to be evaluated time efficiently to maximize the treatment effect. A computational method will be presented that detects rhythmic and periodic patterns according to the critical care EEG terminology (CCET) of the American Clinical Neurophysiology Society (ACNS). The aim is to show that these detected patterns support EEG experts in writing neurophysiological reports. MATERIALS AND METHODS: First of all, three case reports exemplify the evaluation procedure using graphically presented detections. Second, 187 hours of EEG from 10 critical care patients were used in a comparative trial study. For each patient the result of a review session using the EEG and the visualized pattern detections was compared to the original neurophysiology report. RESULTS: In three out of five patients with reported seizures, all seizures were reported correctly. In two patients, several subtle clinical seizures with unclear EEG correlation were missed. Lateralized periodic patterns (LPD) were correctly found in 2/2 patients and EEG slowing was correctly found in 7/9 patients. In 8/10 patients, additional EEG features were found including LPDs, EEG slowing, and seizures. CONCLUSION: The use of automatic pattern detection will assist in review of EEG and increase efficiency. The implementation of bedside surveillance devices using our detection algorithm appears to be feasible and remains to be confirmed in further multicenter studies.

Prediction of rhythmic and periodic EEG patterns and seizures on continuous EEG with early epileptiform discharges.

BACKGROUND: Continuous EEG (cEEG) is necessary to document nonconvulsive seizures (NCS), nonconvulsive status epilepticus (NCSE), as well as rhythmic and periodic EEG patterns of 'ictal-interictal uncertainty' (RPPIIU) including periodic discharges, rhythmic delta activity, and spike-and-wave complexes in neurological intensive care patients. However, cEEG is associated with significant recording and analysis efforts. Therefore, predictors from short-term routine EEG with a reasonably high yield are urgently needed in order to select patients for evaluation with cEEG. OBJECTIVE: The aim of this study was to assess the prognostic significance of early epileptiform discharges (i.e., within the first 30 min of EEG recording) on the following: (1) incidence of ictal EEG patterns and RPPIIU on subsequent cEEG, (2) occurrence of acute convulsive seizures during the ICU stay, and (3) functional outcome after 6 months of follow-up. METHODS: We conducted a separate analysis of the first 30 min and the remaining segments of prospective cEEG recordings according to the ACNS Standardized Critical Care EEG Terminology as well as NCS criteria and review of clinical data of 32 neurological critical care patients. RESULTS: In 17 patients with epileptiform discharges within the first 30 min of EEG (group 1), electrographic seizures were observed in 23.5% (n = 4), rhythmic or periodic EEG patterns of 'ictal-interictal uncertainty' in 64.7% (n = 11), and neither electrographic seizures nor RPPIIU in 11.8% (n = 2). In 15 patients with no epileptiform discharges in the first 30 min of EEG (group 2), no electrographic seizures were recorded on subsequent cEEG, RPPIIU were seen in 26.7% (n = 4), and neither electrographic seizures nor RPPIIU in 73.3% (n = 11). The incidence of EEG patterns on cEEG was significantly different between the two groups (p = 0.008). Patients with early epileptiform discharges developed acute seizures more frequently than patients without early epileptiform discharges (p = 0.009). Finally, functional outcome six months after discharge was significantly worse in patients with early epileptiform discharges (p=0.01). CONCLUSIONS: Epileptiform discharges within the first 30 min of EEG recording are predictive for the occurrence of ictal EEG patterns and for RPPIIU on subsequent cEEG, for acute convulsive seizures during the ICU stay, and for a worse functional outcome after 6 months of follow-up. This article is part of a Special Issue entitled Status Epilepticus.

Prospective assessment and validation of rhythmic and periodic pattern detection in NeuroTrend: A new approach for screening continuous EEG in the intensive care unit.

BACKGROUND: NeuroTrend is a computational method that analyzes long-term scalp EEGs in the ICU according to ACNS standardized critical care EEG terminology (CCET) including electrographic seizures. At present, it attempts to become a screening aid for continuous EEG (cEEG) recordings in the ICU to facilitate the review process and optimize resources. METHODS: A prospective multicenter study was performed in two neurological ICUs including 68 patients who were subjected to video-cEEG. Two reviewers independently annotated the first minute of each hour in the cEEG according to CCET. These segments were also screened for faster patterns with frequencies higher than 4 Hz. The matching annotations (2911 segments) were then used as gold standard condition to test sensitivity and specificity of the rhythmic and periodic pattern detection of NeuroTrend. RESULTS: Interrater agreement showed substantial agreement for localization (main term 1) and pattern type (main term 2) of the CCET. The overall detection sensitivity of NeuroTrend was 94% with high detection rates for periodic discharges (PD = 80%) and rhythmic delta activity (RDA = 82%). Overall specificity was moderate (67%) mainly because of false positive detections of RDA in cases of general slowing. In contrast, a detection specificity of 88% for PDs was reached. Localization revealed only a slight agreement between reviewers and NeuroTrend. CONCLUSIONS: NeuroTrend might be a suitable screening tool for cEEG in the ICU and has the potential to raise efficiency of long-term EEG monitoring in the ICU. At this stage, pattern localization and differentiation between RDA and general slowing need improvement. This article is part of a Special Issue entitled "Status Epilepticus".

Prospective multi-center study of an automatic online seizure detection system for epilepsy monitoring units.

OBJECTIVE: A method for automatic detection of epileptic seizures in long-term scalp-EEG recordings called EpiScan will be presented. EpiScan is used as alarm device to notify medical staff of epilepsy monitoring units (EMUs) in case of a seizure. METHODS: A prospective multi-center study was performed in three EMUs including 205 patients. A comparison between EpiScan and the Persyst seizure detector on the prospective data will be presented. In addition, the detection results of EpiScan on retrospective EEG data of 310 patients and the public available CHB-MIT dataset will be shown. RESULTS: A detection sensitivity of 81% was reached for unequivocal electrographic seizures with false alarm rate of only 7 per day. No statistical significant differences in the detection sensitivities could be found between the centers. The comparison to the Persyst seizure detector showed a lower false alarm rate of EpiScan but the difference was not of statistical significance. CONCLUSIONS: The automatic seizure detection method EpiScan showed high sensitivity and low false alarm rate in a prospective multi-center study on a large number of patients. SIGNIFICANCE: The application as seizure alarm device in EMUs becomes feasible and will raise the efficiency of video-EEG monitoring and the safety levels of patients.

Automatic optimization of parameters for seizure detection systems.

A parameter optimization method for an automatic seizure detection algorithm using the Nelder Mead algorithm is presented. A suitable cost function for joint optimization of sensitivity and false alarm rate is proposed. The optimization is done using EEG datasets from 23 patients and validated on datasets from another 23 patients. The resulting sensitivity was 82.3% with a false alarm rate of 0.24 FA/h. This is a reduction of the false alarm rate by 1.58 FA/h with an acceptable loss of sensitivity of 4.3%.

Combining time series and frequency domain analysis for a automatic seizure detection.

The detection of epileptic seizures in long-term electroencephalographic (EEG) recordings is a time-consuming and tedious task requiring specially trained medical experts. The EpiScan seizure detection algorithm developed by the Austrian Institute of Technology (AIT) has proven to achieve high detection performance with a robust false alarm rate in the clinical setting. This paper introduces a novel time domain method for detection of epileptic seizure patterns with focus on irregular and distorted rhythmic activity. The method scans the EEG for sequences of similar epileptiform discharges and uses a combination of duration and similarity measure to decide for a seizure. The resulting method was tested on an EEG database with 275 patients including over 22000h of unselected and uncut EEG recording and 623 seizures. Used in combination with the EpiScan algorithm we increased the overall sensitivity from 70% to 73% while reducing the false alarm rate from 0.33 to 0.30 alarms per hour.

Automatic seizure detection based on the activity of a set of current dipoles: first steps.

In this paper we show advantages of using an advanced montage scheme with respect to the performance of automatic seizure detection systems. The main goal is to find the best performing montage scheme for our automatic seizure detection system. The new virtual montage is a fix set of dipoles within the brain. The current density signals for these dipoles are derived from the scalp EEG signals based on a smart linear transformation. The reason for testing an alternative approach is that traditional montages (reference, bipolar) have some limitations, e.g. the detection performance depends on the choice of the reference electrode and an extraction of spatial information is often demanding. In this paper we explain the detailed setup of how to adapt a modern seizure detection system to use current density signals. Furthermore, we show results concerning the detection performance of different montage schemes and their combination.

Automatic detection of the seizure onset zone based on ictal EEG.

In this paper we show a proof of concept for novel automatic seizure onset zone detector. The proposed approach utilizes the Austrian Institute of Technology (AIT) seizure detection system EpiScan extended by a frequency domain source localization module. EpiScan was proven to detect rhythmic epileptoform seizure activity often seen during the early phase of epileptic seizures with reasonable high sensitivity and specificity. Additionally, the core module of EpiScan provides complex coefficients and fundamental frequencies representing the rhythmic activity of the ictal EEG signal. These parameters serve as input to a frequency domain version of the Minimum Variance Beamformer to estimate the most dominant source. The position of this source is the detected seizure onset zone. The results are compared to a state of the art wavelet transformation approach based on a manually chosen frequency band. Our first results are encouraging since they coincide with those obtained with the wavelet approach and furthermore show excellent accordance with the medical report for the majority of analyzed seizures. In contrast to the wavelet approach our method has the advantage that it does not rely on a manual selection of the frequency band.