The Individual Ictal Fingerprint: Combining Movement Measures With Ultra Long-Term Subcutaneous EEG in People With Epilepsy

Abstract

Background: Epileptic seizures are caused by abnormal brain wave hypersynchronization leading to a range of signs and symptoms. Tools for detecting seizures in everyday life typically focus on cardiac rhythm, electrodermal activity, or movement (EMG, accelerometry); however, these modalities are not very effective for non-motor seizures. Ultra long-term subcutaneous EEG-devices can detect the electrographic changes that do not depend on clinical changes. Nonetheless, this also means that it is not possible to assess whether a seizure is clinical or subclinical based on an EEG signal alone. Therefore, we combine EEG and movement-related modalities in this work. We focus on whether it is possible to define an individual "multimodal ictal fingerprint" which can be exploited in different epilepsy management purposes. Methods: This study used ultra long-term data from an outpatient monitoring trial of persons with temporal lobe epilepsy obtained with a subcutaneous EEG recording system. Subcutaneous EEG, an EMG estimate and chest-mounted accelerometry were extracted from four persons showing more than 10 well-defined electrographic seizures each. Numerous features were computed from all three modalities. Based on these, the Gini impurity measure of a Random Forest classifier was used to select the most discriminative features for the ictal fingerprint. A total of 74 electrographic seizures were analyzed. Results: The optimal individual ictal fingerprints included features extracted from all three tested modalities: an acceleration component; the power of the estimated EMG activity; and the relative power in the delta (0.5-4 Hz), low theta (4-6 Hz), and high theta (6-8 Hz) bands of the subcutaneous EEG. Multimodal ictal fingerprints were established for all persons, clustering seizures within persons, while separating seizures across persons. Conclusion: The existence of multimodal ictal fingerprints illustrates the benefits of combining multiple modalities such as EEG, EMG, and accelerometry in future epilepsy management. Multimodal ictal fingerprints could be used by doctors to get a better understanding of the individual seizure semiology of people with epilepsy. Furthermore, the multimodal ictal fingerprint gives a better understanding of how seizures manifest simultaneously in different modalities. A knowledge that could be used to improve seizure acknowledgment when reviewing EEG without video.

Keywords: EMG; accelerometry; epilepsy; ictal fingerprint; seizure detection; subcutaneous EEG.

Figure 1

The SubQ solution and its placement in this study. Left: sagittal view of the head showing the placement of implant at the temporal region, recording 2-channel bipolar temporal sqEEG. Right: frontal view of the trunk demonstrating the placement of the external device, recording 3-axis accelerometry of the body trunk.

Figure 2

The multimodal ictal fingerprint. Each color represents a separate participant (blue, orange, green, and red are person B, E, G, and I, respectively). Left: Radar charts of the ictal feature medians and interquartile ranges for each person. F1: X-axis acceleration component; F2: EMG power proximal; F3: EEG relative delta power; F4: EEG relative low theta power; F5: EEG relative high theta power. The heterogeneity across persons is shown. The ictal fingerprint of person B (blue) is dominated by EEG theta power elevation. The fingerprint of person E (orange) has a partial overlap with elevated theta activity. However, the movement component (F1) is a major contributor for this fingerprint. The movement component is also substantial for person I (red), but with no remarkable theta elevation. The ictal fingerprint of person G (green) is not dominated by either of the mentioned features; instead, elevated EMG activity is the main contributor. Right: Pair plot of the first three principal components of the reduced feature space of the ictal periods, where each dot represents a seizure. Scatterplots are shown for each pairing of the principal components, and marginal distributions are plotted along the diagonal (layered kernel density estimates). Within-person clustering and separation across persons are shown. The seizures of each individual could be separated from the rest with an 84.5% accuracy.

Figure 3

Distance-to-ictal-cluster-average vs. distance-to-preictal- cluster-average for all ictal and pre-ictal periods to illustrate the separation capability. Distances are to person-specific cluster averages, calculated as Euclidean norms in the reduced feature space. The solid black line represents the class separation. The separation accuracy was 83.8%. Period is shown with different colors (blue are pre-ictal periods and orange are ictal periods) and the subject ID is shown with different marker types (circles, crosses, squares and plusses are persons B, E, G, and I, respectively).