Detection of Intracranial Signatures of Interictal Epileptiform Discharges from Concurrent Scalp EEG

Abstract

Interictal epileptiform discharges (IEDs) are transient neural electrical activities that occur in the brain of patients with epilepsy. A problem with the inspection of IEDs from the scalp electroencephalogram (sEEG) is that for a subset of epileptic patients, there are no visually discernible IEDs on the scalp, rendering the above procedures ineffective, both for detection purposes and algorithm evaluation. On the other hand, intracranially placed electrodes yield a much higher incidence of visible IEDs as compared to concurrent scalp electrodes. In this work, we utilize concurrent scalp and intracranial EEG (iEEG) from a group of temporal lobe epilepsy (TLE) patients with low number of scalp-visible IEDs. The aim is to determine whether by considering the timing information of the IEDs from iEEG, the resulting concurrent sEEG contains enough information for the IEDs to be reliably distinguished from non-IED segments. We develop an automatic detection algorithm which is tested in a leave-subject-out fashion, where each test subject's detection algorithm is based on the other patients' data. The algorithm obtained a [Formula: see text] accuracy in recognizing scalp IED from non-IED segments with [Formula: see text] accuracy when trained and tested on the same subject. Also, it was able to identify nonscalp-visible IED events for most patients with a low number of false positive detections. Our results represent a proof of concept that IED information for TLE patients is contained in scalp EEG even if they are not visually identifiable and also that between subject differences in the IED topology and shape are small enough such that a generic algorithm can be used.

Deep learning for epileptic intracranial EEG data

“…Although shallow learning of CNN1 yielded poorer results than those in our previous work [13], deep learning of CNN2 had similar performances as in [13]. This was confirmed by McNemars statistical test to assess the significance difference between the CNN1 method and the other three.…”

“…TD-CNN2 and CNN1-CNN2 had significant difference with p < 0.01, while TF-CNN2 was not significant. In other words, our approach CCN2 provided similar performance results as those in our previous work [13]. Yet, the advantage of the methodology proposed herein was to circumvent the use of time-frequency analysis, facilitating the interpretation of the EEG data.…”

“…The parameters for the two CNNs are summarised in Table 2. We compare the proposed CNN architecture with a state-of-the-art classification algorithm [13] that only uses a single logistic regression layer but with handcrafted optimised features. State-of-the-art performance is achieved with TF features (see Section 2) with a logistic regression classifier.…”

“…Many classification algorithms have been developed for the purpose of distinguishing IED from non-IED activity. A review of algorithms is provided in [12,13]. The common denominator of all these methods lies in the statistical description of an IED signal, which can be obtained either through modeling or through the use of a similarity-based algorithm.…”

Detection of Interictal Discharges With Convolutional Neural Networks Using Discrete Ordered Multichannel Intracranial EEG

“…These were evaluated in the background context as described in [35], and following the standard definitions for epileptiform pattern, spike and sharp wave of the International Federation of Clinical Neurophysiology [36] and currently accepted EEG descriptions were taken into account for scoring. Each trial was 65 sample long which is equivalent to 325ms and was given a certainty score (0 − 4) and categorised to one of the following: 0: Non-physiological and physiological artifacts, physiological 'sharpened/spiky' activities (vertex waves, Kcomplexes), and low amplitude irregularities barely distinguishable from the background activity and restricted to 1 − 2 channels.…”