Computer-assisted interpretation of the EEG background pattern: a clinical evaluation

Expert level of detection of interictal discharges with a deep neural network

OBJECTIVE

Deep learning methods have shown potential in automating the detection of interictal epileptiform discharges (IEDs) in electroencephalography (EEG). We compared IED detection using our previously trained deep neural network with a group of experts to assess its potential applicability.

METHODS

First, we performed clinical validation on an internal data set. Seven experts reviewed all EEG studies. Performance agreement between experts and the network was compared at both the EEG and IED levels. All EEG recordings were also processed with Persyst. Subsequently, we performed external validation, with data from four centers, using a hybrid approach, where detections by the deep neural network were reviewed by an expert. In case of disagreement with the original report, the EEG recording was annotated independently by five experts.

RESULTS

For internal validation we included 22 EEG studies with IEDs and 28 EEG studies from controls. At the EEG level, our network showed performance similar to that of the experts. For individual IED detection, the sensitivities between experts ranged from 20.7%-86.4%, whereas the sensitivity of our network was 82.5% (confidence interval [CI]: 77.7%-87.4%) at 99% specificity and a false detection rate (FDR) of <.2/min, outperforming Persyst, with 64.6% sensitivity (CI: 61.4%-67.9%) at 98% specificity. External validation in 174 EEG studies demonstrated that all 85 EEG recordings classified as normal in the original report were classified correctly, with an FDR of .10/min. Of the 89 EEG studies with IEDs according to the report, 56 were correctly classified (Cohen's κ = .62). Visual analysis of the remaining 33 EEG recordings showed high interobserver variability among the five experts (Fleiss' κ = .13).

SIGNIFICANCE

Our deep neural network detects IEDs on par with clinical experts. The external validation in a hybrid approach showed substantial agreement with the original report. Disagreement was due mainly to high interobserver variability. Our deep neural network may support visual EEG analysis and assist in diagnostics, particularly when human resources are limited.

Unveiling variability: A systematic review of reproducibility in visual EEG analysis, with focus on seizures

OBJECTIVE

Reproducibility is key for diagnostic tests involving subjective evaluation by experts. Our aim was to systematically review the reproducibility of visual analysis in clinical electroencephalogram (EEG). In this paper, we give data on the scope of EEG features found, and detailed reproducibility data for the most studied feature.

METHODS

We searched four databases for articles reporting reproducibility in clinical EEG, until June 2023. Two raters screened 24 553 citations, and then 2736 full texts. Quality was assessed according to the GRRAS guidelines.

RESULTS

We found 275 studies (268 interrater and 20 intrarater), addressing 606 different EEG features. Only 38 EEG features had been studied in >2 studies. Most studies had <50 patients and EEGs. The most often addressed feature was seizure detection (62 papers). Interrater reproducibility of seizure detection was substantial-to-almost-perfect with experienced raters and raw EEG (kappa .62-.88). With experienced raters and transformed EEG, reproducibility was substantial (kappa .63-.70). Inexperienced raters had lower reproducibility. Seizure lateralization reproducibility was moderate to substantial (kappa .58-.77) but lower than for seizure detection.

SIGNIFICANCE

Most EEG reproducibility studies are done only once. Intrarater studies are rare. The reproducibility of visual EEG analysis is variable. Interrater reproducibility for seizure detection is substantial-to-perfect with experienced raters and raw EEG, less with inexperienced raters or transformed EEG. The results of visual EEG analysis vary within the same rater, and between raters. There is a need for larger collaborative studies, using improved methodology, as well as more intrarater studies of EEG interpretation.

Scorepochs: A Computer-Aided Scoring Tool for Resting-State M/EEG Epochs

M/EEG resting-state analysis often requires the definition of the epoch length and the criteria in order to select which epochs to include in the subsequent steps. However, the effects of epoch selection remain scarcely investigated and the procedure used to (visually) inspect, label, and remove bad epochs is often not documented, thereby hindering the reproducibility of the reported results. In this study, we present Scorepochs, a simple and freely available tool for the automatic scoring of resting-state M/EEG epochs that aims to provide an objective method to aid M/EEG experts during the epoch selection procedure. We tested our approach on a freely available EEG dataset containing recordings from 109 subjects using the BCI2000 64 channel system.

Machine learning for detection of interictal epileptiform discharges

The electroencephalogram (EEG) is a fundamental tool in the diagnosis and classification of epilepsy. In particular, Interictal Epileptiform Discharges (IEDs) reflect an increased likelihood of seizures and are routinely assessed by visual analysis of the EEG. Visual assessment is, however, time consuming and prone to subjectivity, leading to a high misdiagnosis rate and motivating the development of automated approaches. Research towards automating IED detection started 45 years ago. Approaches range from mimetic methods to deep learning techniques. We review different approaches to IED detection, discussing their performance and limitations. Traditional machine learning and deep learning methods have yielded the best results so far and their application in the field is still growing. Standardization of datasets and outcome measures is necessary to compare models more objectively and decide which should be implemented in a clinical setting.

Efficient use of clinical EEG data for deep learning in epilepsy

OBJECTIVE

Automating detection of Interictal Epileptiform Discharges (IEDs) in electroencephalogram (EEG) recordings can reduce the time spent on visual analysis for the diagnosis of epilepsy. Deep learning has shown potential for this purpose, but the scarceness of expert annotated data creates a bottleneck in the process.

METHODS

We used EEGs from 50 patients with focal epilepsy, 49 patients with generalized epilepsy (IEDs were visually labeled by experts) and 67 controls. The data was filtered, downsampled and cut into two second epochs. We increased the number of input samples containing IEDs through temporal shifting and using different montages. A VGG C convolutional neural network was trained to detect IEDs.

RESULTS

Using the dataset with more samples, we reduced the false positive rate from 2.11 to 0.73 detections per minute at the intersection of sensitivity and specificity. Sensitivity increased from 63% to 96% at 99% specificity. The model became less sensitive to the position of the IED in the epoch and montage.

CONCLUSIONS

Temporal shifting and use of different EEG montages improves performance of deep neural networks in IED detection.

SIGNIFICANCE

Dataset augmentation can reduce the need for expert annotation, facilitating the training of neural networks, potentially leading to a fundamental shift in EEG analysis.

Adult EEG

After more than 85 years of development and use in clinical practice, the electroencephalogram (EEG) remains a dependable, inexpensive, and useful diagnostic tool for the investigation of the electrophysiologic activity of the brain. The advent of digital technology has led to greater sophistication and multiple software applications to extend the utility of EEG beyond the confines of the laboratory. Despite the discovery of new waveforms, basic neurophysiologic principles remain essential to the clinical care of patients. Patterns in the interictal EEG make it possible to clarify the differential diagnosis of paroxysmal neurological events, classify seizure type and epilepsy syndromes, and characterize and quantify seizures when ictal recordings are obtained. EEG can also demonstrate cerebral dysfunction when structural imaging is normal to detect focal or lateralized abnormalities in patients with encephalopathy. High-density EEG with electrical source imaging has improved localization in candidates for epilepsy surgery. Quantitative EEG and broadband EEG are advancing our understanding of the functional processes of the brain itself.

Visual EEG reviewing times with SCORE EEG

•

There is concern that the SCORE reporting standard for EEG takes too long.

•

This study shows that a normal EEG can typically be reported in SCORE EEG in 8 min.

•

Reviewing time is higher for abnormal recordings, and declined by 25% in this study.

Objective

Visual EEG analysis is the gold standard for clinical EEG interpretation and analysis, but there is no published data on how long it takes to review and report an EEG in clinical routine. Estimates of reporting times may inform workforce planning and automation initiatives for EEG. The SCORE standard has recently been adopted to standardize clinical EEG reporting, but concern has been expressed about the time spent reporting.

Methods

Elapsed times were extracted from 5889 standard and sleep-deprived EEGs reported between 2015 and 2017 reported using the SCORE EEG software.

Results

The median review time for standard EEG was 12.5 min, and for sleep deprived EEG 20.9 min. A normal standard EEG had a median review time of 8.3 min. Abnormal EEGs took longer than normal EEGs to review, and had more variable review times. 99% of EEGs were reported within 24 h of end of recording. Review times declined by 25% during the study period.

Conclusion

Standard and sleep-deprived EEG review and reporting times with SCORE EEG are reasonable, increasing with increasing EEG complexity and decreasing with experience. EEG reports can be provided within 24 h.

Significance

Clinical standard and sleep-deprived EEG reporting with SCORE EEG has acceptable reporting times.

Predicting sex from brain rhythms with deep learning

We have excellent skills to extract sex from visual assessment of human faces, but assessing sex from human brain rhythms seems impossible. Using deep convolutional neural networks, with unique potential to find subtle differences in apparent similar patterns, we explore if brain rhythms from either sex contain sex specific information. Here we show, in a ground truth scenario, that a deep neural net can predict sex from scalp electroencephalograms with an accuracy of >80% (p < 10-5), revealing that brain rhythms are sex specific. Further, we extracted sex-specific features from the deep net filter layers, showing that fast beta activity (20-25 Hz) and its spatial distribution is a main distinctive attribute. This demonstrates the ability of deep nets to detect features in spatiotemporal data unnoticed by visual assessment, and to assist in knowledge discovery. We anticipate that this approach may also be successfully applied to other specialties where spatiotemporal data is abundant, including neurology, cardiology and neuropsychology.