Development of Expert-Level Automated Detection of Epileptiform Discharges During Electroencephalogram Interpretation

Abstract

discharges (IEDs) in electroencephalograms (EEGs) are a biomarker of epilepsy, seizure risk, and clinical decline. However, there is a scarcity of experts qualified to interpret EEG results. Prior attempts to automate IED detection have been limited by small samples and have not demonstrated expert-level performance. There is a need for a validated automated method to detect IEDs with expert-level reliability.OBJECTIVE To develop and validate a computer algorithm with the ability to identify IEDs as reliably as experts and classify an EEG recording as containing IEDs vs no IEDs.DESIGN, SETTING, AND PARTICIPANTS A total of 9571 scalp EEG records with and without IEDs were used to train a deep neural network (SpikeNet) to perform IED detection. Independent training and testing data sets were generated from 13 262 IED candidates, independently annotated by 8 fellowship-trained clinical neurophysiologists, and 8520 EEG records containing no IEDs based on clinical EEG reports. Using the estimated spike probability, a classifier designating the whole EEG recording as positive or negative was also built.MAIN OUTCOMES AND MEASURES SpikeNet accuracy, sensitivity, and specificity compared with fellowship-trained neurophysiology experts for identifying IEDs and classifying EEGs as positive or negative or negative for IEDs. Statistical performance was assessed via calibration error and area under the receiver operating characteristic curve (AUC). All performance statistics were estimated using 10-fold cross-validation.RESULTS SpikeNet surpassed both expert interpretation and an industry standard commercial IED detector, based on calibration error (SpikeNet, 0.041; 95% CI, 0.033-0.049; vs industry standard, 0.066; 95% CI, 0.060-0.078; vs experts, mean, 0.183; range, 0.081-0.364) and binary classification performance based on AUC (SpikeNet, 0.980; 95% CI, 0.977-0.984; vs industry standard, 0.882; 95% CI, 0.872-0.893). Whole EEG classification had a mean calibration error of 0.126 (range, 0.109-0.1444) vs experts (mean, 0.197; range, 0.099-0.372) and AUC of 0.847 (95% CI, 0.830-0.865). CONCLUSIONS AND RELEVANCEIn this study, SpikeNet automatically detected IEDs and classified whole EEGs as IED-positive or IED-negative. This may be the first time an algorithm has been shown to exceed expert performance for IED detection in a representative sample of EEGs and may thus be a valuable tool for expedited review of EEGs.

Persistent abnormalities in Rolandic thalamocortical white matter circuits in childhood epilepsy with centrotemporal spikes

“…43 In its validation study, the SpikeNet algorithm was directly compared to neurophysiology-trained experts and Persyst, exceeding both in performance. 34 Here, we detected a possible relationship between spike rate and CI in hemispheres with active spikes using SpikeNet, although this possible finding requires validation in a second dataset. Finally, although spike rate has been observed to remain stable over periods of sleep and the course of weeks, it is possible that our data sample was underpowered to identify a small effect.…”

“…following standard criteria 30 ; (2) using a validated commercial automated wavelet-based detector, Persyst 31,32 ; and (3) using SpikeNet, a validated automated spike detector developed at our institution. 33,34 For each method, all available non-REM sleep epochs were included (mean = 15.2 minutes, range = 4-48 minutes), and a spike rate for each hemisphere was calculated. To minimize artifactual detections, only spikes detected as maximal in the central and temporal channels (consistent with classic CECTS neurophysiology) were included.…”

Machine learning and wearable devices of the future

“…In a clinical environment, deep learning was found to be robust for automated review and quantification of epileptic discharges in patients with generalized epilepsy 44,65 . Another recently published large‐scale study, also used a deep learning–based detection algorithm for epileptiform EEG discharges that was validated against scorings of experts, with remarkable results 43 …”

New era of personalised epilepsy management

“…Machine learning is being explored in epilepsy to forecast and detect seizures through recognition of electroencephalography (EEG) patterns. A recent study used 9571 routinely collected scalp EEG records to train a deep neural network that outperformed experts in detecting interictal epileptiform discharges 11. Researchers have also used time series based algorithms (for example, the line length algorithm used in responsive neurostimulation systems12) to analyse controlled, continuously acquired, intracranial EEG signals to develop seizure warning systems 13.…”