Relationship between high-frequency oscillations and spikes in a case of temporal lobe epilepsy

Epileptic EEG patterns recognition through machine learning techniques and relevant time-frequency features

OBJECTIVES

The present study is designed to explore the process of epileptic patterns' automatic detection, specifically, epileptic spikes and high-frequency oscillations (HFOs), via a selection of machine learning (ML) techniques. The primary motivation for conducting such a research lies mainly in the need to investigate the long-term electroencephalography (EEG) recordings' visual examination process, often considered as a time-consuming and potentially error-prone procedure, requiring a great deal of mental focus and highly experimented neurologists. On attempting to resolve such a challenge, a number of state-of-the-art ML algorithms have been evaluated and compare in terms of performance, to pinpoint the most effective algorithm fit for accurately extracting epileptic EEG patterns.

CONTENT

Based on intracranial as well as simulated EEG data, the attained findings turn out to reveal that the randomforest (RF) method proved to be the most consistently effective approach, significantly outperforming the entirety of examined methods in terms of EEG recordings epileptic-pattern identification. Indeed, the RF classifier appeared to record an average balanced classification rate (BCR) of 92.38 % in regard to spikes recognition process, and 78.77 % in terms of HFOs detection.

SUMMARY

Compared to other approaches, our results provide valuable insights into the RF classifier's effectiveness as a powerful ML technique, fit for detecting EEG signals born epileptic bursts.

OUTLOOK

As a potential future work, we envisage to further validate and sustain our major reached findings through incorporating a larger EEG dataset. We also aim to explore the generative adversarial networks (GANs) application so as to generate synthetic EEG signals or combine signal generation techniques with deep learning approaches. Through this new vein of thought, we actually preconize to enhance and boost the automated detection methods' performance even more, thereby, noticeably enhancing the epileptic EEG pattern recognition area.

Interictal spikes with and without high-frequency oscillation have different single-neuron correlates

Interictal epileptiform discharges (IEDs) are a widely used biomarker in patients with epilepsy but lack specificity. It has been proposed that there are truly epileptogenic and less pathological or even protective IEDs. Recent studies suggest that highly pathological IEDs are characterized by high-frequency oscillations (HFOs). Here, we aimed to dissect these 'HFO-IEDs' at the single-neuron level, hypothesizing that the underlying mechanisms are distinct from 'non-HFO-IEDs'. Analysing hybrid depth electrode recordings from patients with temporal lobe epilepsy, we found that single-unit firing rates were higher in HFO- than in non-HFO-IEDs. HFO-IEDs were characterized by a pronounced pre-peak increase in firing, which coincided with the preferential occurrence of HFOs, whereas in non-HFO-IEDs, there was only a mild pre-peak increase followed by a post-peak suppression. Comparing each unit's firing during HFO-IEDs to its baseline activity, we found many neurons with a significant increase during the HFO component or ascending part, but almost none with a decrease. No such imbalance was observed during non-HFO-IEDs. Finally, comparing each unit's firing directly between HFO- and non-HFO-IEDs, we found that most cells had higher rates during HFO-IEDs and, moreover, identified a distinct subset of neurons with a significant preference for this IED subtype. In summary, our study reveals that HFO- and non-HFO-IEDs have different single-unit correlates. In HFO-IEDs, many neurons are moderately activated, and some participate selectively, suggesting that both types of increased firing contribute to highly pathological IEDs.