Dataset Information

Building an Open Source Classifier for the Neonatal EEG Background: A Systematic Feature-Based Approach From Expert Scoring to Clinical Visualization.

ABSTRACT: Neonatal brain monitoring in the neonatal intensive care units (NICU) requires a continuous review of the spontaneous cortical activity, i.e., the electroencephalograph (EEG) background activity. This needs development of bedside methods for an automated assessment of the EEG background activity. In this paper, we present development of the key components of a neonatal EEG background classifier, starting from the visual background scoring to classifier design, and finally to possible bedside visualization of the classifier results. A dataset with 13,200 5-minute EEG epochs (8-16 channels) from 27 infants with birth asphyxia was used for classifier training after scoring by two independent experts. We tested three classifier designs based on 98 computational features, and their performance was assessed with respect to scoring system, pre- and post-processing of labels and outputs, choice of channels, and visualization in monitor displays. The optimal solution achieved an overall classification accuracy of 97% with a range across subjects of 81-100%. We identified a set of 23 features that make the classifier highly robust to the choice of channels and missing data due to artefact rejection. Our results showed that an automated bedside classifier of EEG background is achievable, and we publish the full classifier algorithm to allow further clinical replication and validation studies.

SUBMITTER: Montazeri S

PROVIDER: S-EPMC8200402 | biostudies-literature | 2021

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Building an Open Source Classifier for the Neonatal EEG Background: A Systematic Feature-Based Approach From Expert Scoring to Clinical Visualization.

Montazeri Saeed S Pinchefsky Elana E Tse Ilse I Marchi Viviana V Kohonen Jukka J Kauppila Minna M Airaksinen Manu M Tapani Karoliina K Nevalainen Päivi P Hahn Cecil C Tam Emily W Y EWY Stevenson Nathan J NJ Vanhatalo Sampsa S

Frontiers in human neuroscience 20210531

Neonatal brain monitoring in the neonatal intensive care units (NICU) requires a continuous review of the spontaneous cortical activity, i.e., the electroencephalograph (EEG) background activity. This needs development of bedside methods for an automated assessment of the EEG background activity. In this paper, we present development of the key components of a neonatal EEG background classifier, starting from the visual background scoring to classifier design, and finally to possible bedside vis ...[more]

PMID: 34135744

Similar Datasets

Project description:PurposeMany neonates undergo electroencephalogram (EEG) monitoring to identify and manage acute symptomatic seizures. Information about brain function contained in the EEG background data may also help predict neurobehavioral outcomes. For EEG background features to be useful as prognostic indicators, the interpretation of these features must be standardized across electroencephalographers. We aimed at determining the interrater and intrarater agreement among electroencephalographers interpreting neonatal EEG background patterns.MethodsFive neonatal electroencephalographers reviewed 5-to-7.5-minute epochs of EEG from full-term neonates who underwent continuous conventional EEG monitoring. The EEG assessment tool used to classify background patterns was based on the American Clinical Neurophysiology Society's guideline for neonatal EEG terminology. Interrater and intrarater agreement were measured using Kappa coefficients.ResultsInterrater agreement was consistently highest for voltage (binary: substantial, kappa = 0.783; categorical: moderate, kappa = 0.562), seizure presence (fair-substantial; kappa = 0.375-0.697), continuity (moderate; kappa = 0.481), burst voltage (moderate; kappa = 0.574), suppressed background presence (moderate-substantial; kappa = 0.493-0.643), delta activity presence (fair-moderate; kappa = 0.369-0.432), theta activity presence (fair-moderate; kappa = 0.347-0.600), presence of graphoelements (fair; kappa = 0.381), and overall impression (binary: moderate, kappa = 0.495; categorical: fair-moderate, kappa = 0.347, 0.465). Agreement was poor or inconsistent for all other patterns. Intrarater agreement was variable, with highest average agreement for voltage (binary: substantial, kappa = 0.75; categorical: substantial, kappa = 0.714) and highest consistent agreement for continuity (moderate-substantial; kappa = 0.43-0.67) and overall impression (moderate-substantial; kappa = 0.42-0.68).ConclusionsThis study demonstrates substantial variability in neonatal EEG background interpretation across electroencephalographers, indicating a need for educational and technological strategies aimed at improving performance.

Project description:A key feature of normal neonatal EEG at term age is interhemispheric synchrony (IHS), which refers to the temporal co-incidence of bursting across hemispheres during trace alternant EEG activity. The assessment of IHS in both clinical and scientific work relies on visual, qualitative EEG assessment without clearly quantifiable definitions. A quantitative measure, activation synchrony index (ASI), was recently shown to perform well as compared to visual assessments. The present study was set out to test whether IHS is stable enough for clinical use, and whether it could be an objective feature of EEG normality. We analyzed 31 neonatal EEG recordings that had been clinically classified as normal (n = 14) or abnormal (n = 17) using holistic, conventional visual criteria including amplitude, focal differences, qualitative synchrony, and focal abnormalities. We selected 20-min epochs of discontinuous background pattern. ASI values were computed separately for different channel pair combinations and window lengths to define them for the optimal ASI intraindividual stability. Finally, ROC curves were computed to find trade-offs related to compromised data lengths, a common challenge in neonatal EEG studies. Using the average of four consecutive 2.5-min epochs in the centro-occipital bipolar derivations gave ASI estimates that very accurately distinguished babies clinically classified as normal vs. abnormal. It was even possible to draw a cut-off limit (ASI~3.6) which correctly classified the EEGs in 97% of all cases. Finally, we showed that compromising the length of EEG segments from 20 to 5 min leads to increased variability in ASI-based classification. Our findings support the prior literature that IHS is an important feature of normal neonatal brain function. We show that ASI may provide diagnostic value even at individual level, which strongly supports its use in prospective clinical studies on neonatal EEG as well as in the feature set of upcoming EEG classifiers.

Project description:BackgroundIntrapartum-related neonatal encephalopathy (NE) is a leading cause of childhood mortality and morbidity. Continuous electroencephalography (EEG) is gold standard for neonatal brain monitoring; however, low-income country data is lacking. We examined EEG in a Ugandan cohort with NE to describe feasibility, background activity, seizure prevalence and burden, and associations with clinical presentation and outcome.MethodsNeonates with NE were recruited from a single hospital referral centre in Kampala, Uganda (Oct 2019-Oct 2020) and underwent EEG monitoring. Feasibility was assessed as to whether EEG monitoring of diagnostic quality could be achieved from days 1-5. Evolution of clinical presentation was assessed by Sarnat classification and daily Thompson score was performed. EEG background severity was graded at 12, 24, 48 and 72 h after birth, and at time of Thompson score. Seizures were annotated remotely by experts and assessed for frequency, duration, burden, and status epilepticus. Early childhood outcome was assessed at follow up, and adverse outcome defined as death or neurodevelopmental impairment (NDI) at 18-24 months of age.FindingsIn this prospective feasibility cohort study, diagnostic quality EEGs were recorded for 50 of 51 recruited neonates (median duration 71.4 h, IQR 52.4-72.2), indicating feasibility. Of 39 participants followed to 18-24 months, 13 died and 7 had NDI. Daily Thompson score and EEG background grade were strongly correlated across all timepoints (days 1-5). Thompson score of ≥7 was most predictive of moderate-severe EEG background abnormality (AUC 0.83). Prognostic accuracy of moderate-severe EEG background grade to predict NDI was high (AUC 0.74). Electrographic seizures were seen in 52% (26); median seizure burden was high at 264 min (IQR 27.8-523.7, range 1.3-1374.1); half (13) had status epilepticus.InterpretationEEG monitoring was feasible as a research tool in this sub-Saharan Africa setting. EEG background activity correlated strongly with scored neurological assessment and predicted adverse early childhood outcome. Seizure prevalence and burden, including status epilepticus, were high in this uncooled cohort with important potential longer-term implications for survivors.FundingBill & Melinda Gates Foundation grant number OPP1210890; Wellcome Trust Innovator award (209325/Z/17/Z).

Dataset Information

Building an Open Source Classifier for the Neonatal EEG Background: A Systematic Feature-Based Approach From Expert Scoring to Clinical Visualization.

Publications

Building an Open Source Classifier for the Neonatal EEG Background: A Systematic Feature-Based Approach From Expert Scoring to Clinical Visualization.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets