Dataset Information

Seizure identification in the ICU using quantitative EEG displays.

ABSTRACT:

Objective

To evaluate the diagnostic accuracy of 2 quantitative EEG display tools, color density spectral array (CDSA) and amplitude-integrated EEG (aEEG), for seizure identification in the intensive care unit (ICU).

Methods

A set of 27 continuous EEG recordings performed in pediatric ICU patients was transformed into 8-channel CDSA and aEEG displays. Three neurophysiologists underwent 2 hours of training to identify seizures using these techniques. They were then individually presented with a series of CDSA and aEEG displays, blinded to the raw EEG, and asked to mark any events suspected to be seizures. Their performance was compared to seizures identified on the underlying conventional EEG.

Results

The 27 EEG recordings contained 553 discrete seizures over 487 hours. The median sensitivity for seizure identification across all recordings was 83.3% using CDSA and 81.5% using aEEG. However, among individual recordings, the sensitivity ranged from 0% to 100%. Factors reducing the sensitivity included low-amplitude, short, and focal seizures. False-positive rates were generally very low, with misidentified seizures occurring once every 17-20 hours.

Conclusions

Both CDSA and aEEG demonstrate acceptable sensitivity and false-positive rates for seizure identification among critically ill children. Accuracy of these tools would likely improve during clinical use, when findings can be correlated in real-time with the underlying raw EEG. In the hands of neurophysiologists, CDSA and aEEG displays represent useful screening tools for seizures during continuous EEG monitoring in the ICU. The suitability of these tools for bedside use by ICU nurses and physicians requires further study.

SUBMITTER: Stewart CP

PROVIDER: S-EPMC2974462 | biostudies-literature | 2010 Oct

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Seizure identification in the ICU using quantitative EEG displays.

Stewart C P CP Otsubo H H Ochi A A Sharma R R Hutchison J S JS Hahn C D CD

Neurology 20100922 17

<h4>Objective</h4>To evaluate the diagnostic accuracy of 2 quantitative EEG display tools, color density spectral array (CDSA) and amplitude-integrated EEG (aEEG), for seizure identification in the intensive care unit (ICU).<h4>Methods</h4>A set of 27 continuous EEG recordings performed in pediatric ICU patients was transformed into 8-channel CDSA and aEEG displays. Three neurophysiologists underwent 2 hours of training to identify seizures using these techniques. They were then individually pre ...[more]

PMID: 20861452

Similar Datasets

Project description:Background and objectiveEpilepsy is one of the most common neurological disorders, whose development is typically detected via early seizures. Electroencephalogram (EEG) is prevalently employed for seizure identification due to its routine and low expense collection. The stochastic nature of EEG makes manual seizure inspections laborsome, motivating automated seizure identification. The relevant literature focuses mostly on supervised machine learning. Despite their success, supervised methods require expert labels indicating seizure segments, which are difficult to obtain on clinically-acquired EEG. Thus, we aim to devise an unsupervised method for seizure identification on EEG.MethodsWe propose the first fully-unsupervised deep learning method for seizure identification on raw EEG, using a variational autoencoder (VAE). In doing so, we train the VAE on recordings without seizures. As training captures non-seizure activity, we identify seizures with respect to the reconstruction errors at inference time. Moreover, we extend the traditional VAE training loss to suppress EEG artifacts. Our method does not require ground-truth expert labels indicating seizure segments or manual feature extraction.ResultsWe implement our method using the PyTorch library and execute experiments on an NVIDIA V100 GPU. We evaluate our method on three benchmark EEG datasets: (i) intracranial recordings from the University of Pennsylvania and the Mayo Clinic, (ii) scalp recordings from the Temple University Hospital of Philadelphia, and (iii) scalp recordings from the Massachusetts Institute of Technology and the Boston Children's Hospital. To assess performance, we report accuracy, precision, recall, Area under the Receiver Operating Characteristics Curve (AUC), and p-value under the Welch t-test for distinguishing seizure vs. non-seizure EEG windows. Our approach can successfully distinguish seizures from non-seizure activity, with up to 0.83 AUC on intracranial recordings. Moreover, our algorithm has the potential for real-time inference, by processing at least 10 s of EEG in a second.ConclusionWe take the first successful steps in deep learning-based unsupervised seizure identification on raw EEG. Our approach has the potential of alleviating the burden on clinical experts regarding laborsome EEG inspections for seizures. Furthermore, aiding the identification of early seizures via our method could facilitate successful detection of epilepsy development and initiate antiepileptogenic therapies.

Project description:BackgroundEpilepsy surgery is a potentially curative treatment option for pharmacoresistent patients. If non-invasive methods alone do not allow to delineate the epileptogenic brain areas the surgical candidates undergo long-term monitoring with intracranial EEG. Visual EEG analysis is then used to identify the seizure onset zone for targeted resection as a standard procedure.MethodsDespite of its great potential to assess the epileptogenicty of brain tissue, quantitative EEG analysis has not yet found its way into routine clinical practice. To demonstrate that quantitative EEG may yield clinically highly relevant information we retrospectively investigated how post-operative seizure control is associated with four selected EEG measures evaluated in the resected brain tissue and the seizure onset zone. Importantly, the exact spatial location of the intracranial electrodes was determined by coregistration of pre-operative MRI and post-implantation CT and coregistration with post-resection MRI was used to delineate the extent of tissue resection. Using data-driven thresholding, quantitative EEG results were separated into normally contributing and salient channels.ResultsIn patients with favorable post-surgical seizure control a significantly larger fraction of salient channels in three of the four quantitative EEG measures was resected than in patients with unfavorable outcome in terms of seizure control (median over the whole peri-ictal recordings). The same statistics revealed no association with post-operative seizure control when EEG channels contributing to the seizure onset zone were studied.ConclusionsWe conclude that quantitative EEG measures provide clinically relevant and objective markers of target tissue, which may be used to optimize epilepsy surgery. The finding that differentiation between favorable and unfavorable outcome was better for the fraction of salient values in the resected brain tissue than in the seizure onset zone is consistent with growing evidence that spatially extended networks might be more relevant for seizure generation, evolution and termination than a single highly localized brain region (i.e. a "focus") where seizures start.

Project description:Seizure prediction might be the solution to tackle the apparent unpredictability of seizures in patients with drug-resistant epilepsy, which comprise about a third of all patients with epilepsy. Designing seizure prediction models involves defining the pre-ictal period, a transition stage between inter-ictal brain activity and the seizure discharge. This period is typically a fixed interval, with some recent studies reporting the evaluation of different patient-specific pre-ictal intervals. Recently, researchers have aimed to determine the pre-ictal period, a transition stage between regular brain activity and a seizure. Authors have been using deep learning models given the ability of such models to automatically perform pre-processing, feature extraction, classification, and handling temporal and spatial dependencies. As these approaches create black-box models, clinicians may not have sufficient trust to use them in high-stake decisions. By considering these problems, we developed an evolutionary seizure prediction model that identifies the best set of features while automatically searching for the pre-ictal period and accounting for patient comfort. This methodology provides patient-specific interpretable insights, which might contribute to a better understanding of seizure generation processes and explain the algorithm's decisions. We tested our methodology on 238 seizures and 3687 h of continuous data, recorded on scalp recordings from 93 patients with several types of focal and generalised epilepsies. We compared the results with a seizure surrogate predictor and obtained a performance above chance for 32% patients. We also compared our results with a control method based on the standard machine learning pipeline (pre-processing, feature extraction, classifier training, and post-processing), where the control marginally outperformed our approach by validating 35% of the patients. In total, 54 patients performed above chance for at least one method: our methodology or the control one. Of these 54 patients, 21 ([Formula: see text]38%) were solely validated by our methodology, while 24 ([Formula: see text]44%) were only validated by the control method. These findings may evidence the need for different methodologies concerning different patients.

Project description:Epilepsy is a chronic brain disease that causes persistent and severe damage to the physical and mental health of patients. Daily effective prediction of epileptic seizures is crucial for epilepsy patients especially those with refractory epilepsy. At present, a large number of deep learning algorithms such as Convolutional Neural Networks and Recurrent Neural Networks have been used to predict epileptic seizures and have obtained better performance than traditional machine learning methods. However, these methods usually transform the Electroencephalogram (EEG) signal into a Euclidean grid structure. The conversion suffers from loss of adjacent spatial information, which results in deep learning models requiring more storage and computational consumption in the process of information fusion after information extraction. This study proposes a general Graph Convolutional Networks (GCN) model architecture for predicting seizures to solve the problem of oversized seizure prediction models based on exploring the graph structure of EEG signals. As a graph classification task, the network architecture includes graph convolution layers that extract node features with one-hop neighbors, pooling layers that summarize abstract node features; and fully connected layers that implement classification, resulting in superior prediction performance and smaller network size. The experiment shows that the model has an average sensitivity of 96.51%, an average AUC of 0.92, and a model size of 15.5 k on 18 patients in the CHB-MIT scalp EEG dataset. Compared with traditional deep learning methods, which require a large number of parameters and computational effort and are demanding in terms of storage space and energy consumption, this method is more suitable for implementation on compact, low-power wearable devices as a standard process for building a generic low-consumption graph network model on similar biomedical signals. Furthermore, the edge features of graphs can be used to make a preliminary determination of locations and types of discharge, making it more clinically interpretable.

Dataset Information

Seizure identification in the ICU using quantitative EEG displays.

Objective

Methods

Results

Conclusions

Publications

Seizure identification in the ICU using quantitative EEG displays.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets