Dataset Information

Deep learning-based automated detection and multiclass classification of focal interictal epileptiform discharges in scalp electroencephalograms.

ABSTRACT: Detection and spatial distribution analyses of interictal epileptiform discharges (IEDs) are important for diagnosing, classifying, and treating focal epilepsy. This study proposes deep learning-based models to detect focal IEDs in electroencephalography (EEG) recordings of the frontal, temporal, and occipital scalp regions. This study included 38 patients with frontal (n = 15), temporal (n = 13), and occipital (n = 10) IEDs and 232 controls without IEDs from a single tertiary center. All the EEG recordings were segmented into 1.5-s epochs and fed into 1- or 2-dimensional convolutional neural networks to construct binary classification models to detect IEDs in each focal region and multiclass classification models to categorize IEDs into frontal, temporal, and occipital regions. The binary classification models exhibited accuracies of 79.3-86.4%, 93.3-94.2%, and 95.5-97.2% for frontal, temporal, and occipital IEDs, respectively. The three- and four-class models exhibited accuracies of 87.0-88.7% and 74.6-74.9%, respectively, with temporal, occipital, and non-IEDs F1-scores of 89.9-92.3%, 84.9-90.6%, and 84.3-86.0%; and 86.6-86.7%, 86.8-87.2%, and 67.8-69.2% for the three- and four-class (frontal, 50.3-58.2%) models, respectively. The deep learning-based models could help enhance EEG interpretation. Although they performed well, the resolution of region-specific focal IED misinterpretations and further model improvement are needed.

SUBMITTER: Chung YG

PROVIDER: S-EPMC10130023 | biostudies-literature | 2023 Apr

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Deep learning-based automated detection and multiclass classification of focal interictal epileptiform discharges in scalp electroencephalograms.

Chung Yoon Gi YG Lee Woo-Jin WJ Na Sung Min SM Kim Hunmin H Hwang Hee H Yun Chang-Ho CH Kim Ki Joong KJ

Scientific reports 20230425 1

Detection and spatial distribution analyses of interictal epileptiform discharges (IEDs) are important for diagnosing, classifying, and treating focal epilepsy. This study proposes deep learning-based models to detect focal IEDs in electroencephalography (EEG) recordings of the frontal, temporal, and occipital scalp regions. This study included 38 patients with frontal (n = 15), temporal (n = 13), and occipital (n = 10) IEDs and 232 controls without IEDs from a single tertiary center. All the EE ...[more]

PMID: 37185941

Similar Datasets

Project description:ImportanceThe validity of using electroencephalograms (EEGs) to diagnose epilepsy requires reliable detection of interictal epileptiform discharges (IEDs). Prior interrater reliability (IRR) studies are limited by small samples and selection bias.ObjectiveTo assess the reliability of experts in detecting IEDs in routine EEGs.Design, setting, and participantsThis prospective analysis conducted in 2 phases included as participants physicians with at least 1 year of subspecialty training in clinical neurophysiology. In phase 1, 9 experts independently identified candidate IEDs in 991 EEGs (1 expert per EEG) reported in the medical record to contain at least 1 IED, yielding 87 636 candidate IEDs. In phase 2, the candidate IEDs were clustered into groups with distinct morphological features, yielding 12 602 clusters, and a representative candidate IED was selected from each cluster. We added 660 waveforms (11 random samples each from 60 randomly selected EEGs reported as being free of IEDs) as negative controls. Eight experts independently scored all 13 262 candidates as IEDs or non-IEDs. The 1051 EEGs in the study were recorded at the Massachusetts General Hospital between 2012 and 2016.Main outcomes and measuresPrimary outcome measures were percentage of agreement (PA) and beyond-chance agreement (Gwet κ) for individual IEDs (IED-wise IRR) and for whether an EEG contained any IEDs (EEG-wise IRR). Secondary outcomes were the correlations between numbers of IEDs marked by experts across cases, calibration of expert scoring to group consensus, and receiver operating characteristic analysis of how well multivariate logistic regression models may account for differences in the IED scoring behavior between experts.ResultsAmong the 1051 EEGs assessed in the study, 540 (51.4%) were those of females and 511 (48.6%) were those of males. In phase 1, 9 experts each marked potential IEDs in a median of 65 (interquartile range [IQR], 28-332) EEGs. The total number of IED candidates marked was 87 636. Expert IRR for the 13 262 individually annotated IED candidates was fair, with the mean PA being 72.4% (95% CI, 67.0%-77.8%) and mean κ being 48.7% (95% CI, 37.3%-60.1%). The EEG-wise IRR was substantial, with the mean PA being 80.9% (95% CI, 76.2%-85.7%) and mean κ being 69.4% (95% CI, 60.3%-78.5%). A statistical model based on waveform morphological features, when provided with individualized thresholds, explained the median binary scores of all experts with a high degree of accuracy of 80% (range, 73%-88%).Conclusions and relevanceThis study's findings suggest that experts can identify whether EEGs contain IEDs with substantial reliability. Lower reliability regarding individual IEDs may be largely explained by various experts applying different thresholds to a common underlying statistical model.

Project description:IntroductionDespite all the efforts for optimizing epilepsy management in children over the past decades, there is no clear consensus regarding whether to treat or not to treat epileptiform discharges (EDs) after a first unprovoked seizure or the optimal duration of therapy with anti-seizure medication (ASM). It is therefore highly needed to find markers on scalp electroencephalogram (EEG) that can help identify pathological EEG discharges that require treatment.Aim of the studyThis retrospective study aimed to identify whether the coexistence of ripples/high-frequency oscillations (HFOs) with interictal EDs (IEDs) in routinely acquired scalp EEG is associated with a higher risk of seizure recurrence and could be used as a prognostic marker.Methods100 children presenting with new onset seizure to Children's Medical Center- Dallas during 2015-2016, who were not on ASM and had focal EDs on an awake and sleep EEG recorded with sample frequency of 500 HZ, were randomly identified by database review. EEGs were analyzed blinded to the data of the patients. HFOs were visually identified using review parameters including expanded time base and adjusted filter settings.ResultsThe average age of patients was 6.3 years (±4.35 SD). HFOs were visually identified in 19% of the studied patients with an inter-rater reliability of 99% for HFO negative discharges and 78% agreement for identification of HFOs. HFOs were identified more often in the younger age group; however, they were identified in 11% of patients >5 years old. They were more frequently associated with spikes than with sharp waves and more often with higher amplitude EDs. Patients with HFOs were more likely to have a recurrence of seizures in the year after the first seizure (P < 0.05) and to continue to have seizures after 2 years (P < 0.0001). There was no statistically significant difference between the two groups with regards to continuing ASM after 2 years.ConclusionIncluding analysis for HFOs in routine EEG interpretation may increase the yield of the study and help guide the decision to either start or discontinue ASM. In the future, this may also help to identify pathological discharges with deleterious effects on the growing brain and set a new target for the management of epilepsy.

Project description:ObjectiveThis study was undertaken to evaluate the influence that subject-specific factors have on intracranial interictal epileptiform discharge (IED) rates in persons with refractory epilepsy.MethodsOne hundred fifty subjects with intracranial electrodes performed multiple sessions of a free recall memory task; this standardized task controlled for subject attention levels. We utilized a dominance analysis to rank the importance of subject-specific factors based on their relative influence on IED rates. Linear mixed-effects models were employed to comprehensively examine factors with highly ranked importance.ResultsAntiseizure medication (ASM) status, time of testing, and seizure onset zone (SOZ) location were the highest-ranking factors in terms of their impact on IED rates. The average IED rate of electrodes in SOZs was 34% higher than the average IED rate of electrodes outside of SOZs (non-SOZ; p < .001). However, non-SOZ electrodes had similar IED rates regardless of the subject's SOZ location (p = .99). Subjects on older generation (p < .001) and combined generation (p < .001) ASM regimens had significantly lower IED rates relative to the group taking no ASMs; newer generation ASM regimens demonstrated a nonsignificant association with IED rates (p = .13). Of the ASMs included in this study, the following ASMs were associated with significant reductions in IED rates: levetiracetam (p < .001), carbamazepine (p < .001), lacosamide (p = .03), zonisamide (p = .01), lamotrigine (p = .03), phenytoin (p = .03), and topiramate (p = .01). We observed a nonsignificant association between time of testing and IED rates (morning-afternoon p = .15, morning-evening p = .85, afternoon-evening p = .26).SignificanceThe current study ranks the relative influence that subject-specific factors have on IED rates and highlights the importance of considering certain factors, such as SOZ location and ASM status, when analyzing IEDs for clinical or research purposes.

Dataset Information

Deep learning-based automated detection and multiclass classification of focal interictal epileptiform discharges in scalp electroencephalograms.

Publications

Deep learning-based automated detection and multiclass classification of focal interictal epileptiform discharges in scalp electroencephalograms.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets