Abstracts

Rationale:
80% of refractory epilepsy patients experience focal seizures, which are often complicated to detect strictly on EEG due to their low occurrence and complex morphology (French JA. Epilepsia 2007;48:3-7). The current standard of seizure logging at home is the subjective and inaccurate seizure diary. The hospital 10-20 video-EEG device remains the gold-standard for accurately measuring and diagnosing seizures; however, its lack of portability largely inhibits patients from wearing these in their daily life, especially for prolonged periods of time. This poses a crucial need for accurate and user-friendly multimodal seizure monitoring devices at home, which are also compatible with clinical workflows in the epilepsy monitoring unit (EMU).
Method:
The Byteflies Sensor Dot can accurately measure 2-channels of biopotentials with motion in various locations of the body. During the preparation of the multicenter SeizeIT2 (NCT04284072) trial, people admitted to the EMU at the University Hospital of Leuven for routine video-EEG monitoring were equipped with at least one Sensor Dot to record two channels of behind-the-ear (BTE) EEG. For patients with focal epilepsy, this set-up consisted of one cross-head channel and one local channel. For some subjects, motion, EMG, and heart rate (HR) were also recorded. All patients were asked to fill out a seizure diary. A total of 31 patients diagnosed with focal epilepsy were recorded. Of these, 6 patients had a total of 9 days of data with seizures visible on the EEG.
Results:
The use of multiple physiologic signals creates a wider and more holistic approach to measuring these complex focal seizures. Figure 1 exhibits a patient outfitted with Sensor Dots measuring EEG, EMG, HR, and motion. The figure shows a clear ictal response in all of the signals during a focal to bilateral tonic-clonic seizure. Figure 2 shows a zoomed in view of the cross-head EEG channel. In other cases, ictal events on non-EEG signals corroborate EEG negative recordings, highlighting the power of multimodal signals to monitor focal seizures with a wearable device. Despite the reduced montages compared to 10-20 EEG, the collected wearable data provides a familiar set of physiologic signals for the epileptologist, compatible with typical workflows. Non-blinded retrospective data review based on known video-EEG annotations of this preliminary focal dataset (n = 6, 45 seizures) resulted in a sensitivity of 94%. Blind review of the Sensor Dot vs. video-EEG and vs. the seizure diary data will occur as part of the SeizeIT2 trial.
Conclusion:
User-friendly wearable devices are needed to replace seizure diaries with objective measurements that replicate the quality and clinical workflows of traditional EMU equipment. A preliminary non-blinded review of focal seizure data measured with the multimodal Byteflies Sensor Dot indicates that focal seizures can be detected with high sensitivity. The data collected as part of the SeizeIT2 trial will be used to rigorously examine  the potential of the Sensor Dot as a home seizure monitoring device for focal and generalized seizures.
Funding:
:Vlaio (SeizeIT project: HBC 2016.0167) EIT Health Grant (project ID 19263)