Key result
Deep neural network detects epileptiform discharges with ~83% sensitivity, matching experts and beating standard software.
Why the study?
Deep learning methods for automated detection of interictal epileptiform discharges in EEGs require validation against expert visual assessment for clinical applicability.
Does a deep neural network detect interictal epileptiform discharges in EEGs with similar performance to human experts?
Observational (n=50)
Single-blind
No
Does a deep neural network detect interictal epileptiform discharges in EEGs with similar performance to human experts?
Absolute Event Rate: 82.5% vs 64.6%
A deep neural network can detect interictal epileptiform discharges in EEGs with high sensitivity and specificity, performing comparably to human experts and potentially supporting clinical visual EEG analysis.
Deep learning methods have shown potential in automating interictal epileptiform discharge (IED) detection in electroencephalograms (EEGs). To implement this in a clinical setting, it needs to have similar performance to visual assessment. We aim to compare a neural network trained for IED detection with a group of experts for validation and assessment of its potential applicability in a clinical setting. We processed EEGs from 20 patients with focal and generalized epilepsy and 30 controls with a neural network. Seven experts labeled the IEDs in the dataset. Kappa scores, sensitivity and specificity were calculated. Network performance was compared to the experts at EEG and at IED level, as well as with an industry standard, Persyst. For EEG level classification, the neural network showed a similar trend to the experts. For individual IED detection, sensitivities between experts ranged from 20.7–86.4%. The sensitivity of our network was 82.5% at 99.0% specificity, and it outperformed Persyst. The network can detect IEDs in agreement with the experts with a high sensitivity and specificity. This indicates that the algorithm can potentially be used in the clinic to support visual EEG analysis or provide access to diagnostics if human resources are limited.
No takes yet. Share an insight, caveat, or question.
Lourenço et al. (2023) conducted an observational in Epilepsy (n=50). Deep neural network (modified VGG C) vs. Visual assessment by seven clinical experts and Persyst software was evaluated on Sensitivity of individual interictal epileptiform discharge (IED) detection at 99% specificity (95% CI 77.7-87.4). A deep neural network detected interictal epileptiform discharges with an average sensitivity of 82.5% at 99.0% specificity, performing on par with clinical experts and outperforming the industry standard Persyst software.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: