README

This dataset is a BIDS compatible version of the SeizeIT2 dataset. It reorganizes the file structure to comply with the BIDS specification. To this effect:

• Metadata was organized according to BIDS.
• Data in the edf files contains wearable EEG, ECG, EMG and movement data recorded with the Sensor Dot device.
• Annotations were formatted as BIDS-score compatible tsv files.

Contact person

The dataset was published by Miguel Bhagubai and Christos Chatzichristos.

Overview

Project name

SeizeIT2 Dataset

Year that the project ran

2024

Description of the dataset

The SeizeIT2 project (clinicaltrials.gov: NCT04284072), a multicenter, prospective study, was carried out to validate the Sensor Dot device in adult and pediatric patients with epilepsy. Participants were included if they had a history of refractory epilepsy and were admitted to the Epilepsy Monitoring Unit (EMU) for long-term vEEG monitoring as a presurgical evaluation procedure. The exclusion criteria included patients with skin conditions or allergies that prevented the placement of the electrodes and adhesives or had implanted devices, such as neurostimulators or pacemakers. All participants provided written informed consent. The data collection started on January 10, 2020, and ended on June 30, 2022. The study was approved by the UZ Leuven ethics committee (approval ID: S63631, ClinicalTrials.gov, NCT04284072), anonymization and sharing of the data was also approved by the same committee (S67350 - amendment 1).

The dataset comprises 125 patients (51 female, 41\%) from 5 different European EMUs: University Hospital Leuven (Belgium), Freiburg University Medical Center (Germany), RWTH University of Aachen (Germany), Karolinska University Hospital (Sweden) and Coimbra University Hospital (Portugal). The University Hospital Leuven was the only center that enrolled pediatric patients. The dataset includes only data from patients with focal epilepsy who experienced one or more seizure episodes during the monitoring period.

Methods

The participants were recorded with the specific center's vEEG monitoring equipment, where the EEG electrodes were placed according to the 10-20 system or the 25-electrode array of the International Federation of Clinical Neurophysiology. The SD device was used to record wearable data simultaneously with the vEEG. The device has a size of 24.5 x 33.5 x 7.73 mm and weighs approximately 6.3 grams. The wearable device measures data at a sampling frequency of 250 Hz and has a battery life of approximately 24 hours. Two recording devices were used: one placed in the patient's upper back using a patch and connected to electrodes attached behind the ear, on the mastoid bone (EEG SD); another placed on the left side of the chest, with two electrodes extended to the lower left rib cage and the fourth intercostal space in the left parasternal position to measure ECG, and two electrodes extended to the left deltoid muscle to measure EMG data (ECG/EMG SD). The module itself contains accelerometers (ACC) and gyroscopes (GYR), which measured movement data at a sampling rate of 25 Hz. The EEG SD electrode placement depended on the patient's medical history and is based on the seizure type and onset. When the seizures were suspected to originate from the left hemisphere, two electrodes were placed on the left side and one on the right side, forming one left same-side channel and one cross-head channel. Analogously, if seizures were suspected to originate from the right hemisphere, the same-side channel was derived from two electrodes placed behind the right ear. The dataset includes patients who were suspected to have generalized seizures (but had focal seizures) as well, and in this case, the cross-head channel was non-existent and replaced by an additional lateral channel by using two electrodes on each ear.

Dataset contents

The complete dataset contains around 11 640 hours of wearable data. Four different modalities were recorded for most participants: bte-EEG, ECG, EMG and movement data. All participants' data within the dataset contain wearable bte-EEG. In 3\% of the dataset, ECG, EMG and movement data were not included due to technical failures or errors in the setup. In total, 886 focal seizures were recorded with the wearable device. The mean duration of the recorded seizures was 58 seconds, ranging between 3 seconds and 16 minutes. The majority of the seizures were focal aware (FA) and focal impaired awareness (FIA), with 317 and 393 occurrences, respectively. From the remaining seizures, 55 were focal-to-bilateral tonic clinic (FBTC), 12 were focal with unclear awareness status, 2 were subclinical focal seizures and 93 had unknown or unreported onset. There was a predominance of seizures with onset on the left hemisphere (44\%). In 12\% of the seizures, the onset was located in the right hemisphere, 1\% had a bilateral onset and in 43\% of the seizures the onset was unclear. Regarding localization, the seizure onsets were distributed over the central, frontal, temporal, occipital, parietal and insula lobes, with a predominance of temporal lobe seizures (30\%). Several seizures recorded could not be paired with a clear onset lobe (26\%).

[The SeizeIT1 dataset will no longer be shared upon request due to expiry of the ethical approval. You can get access to SeizeIT2 if you sign up to the challenge at https://biomedepi.github.io/seizure_detection_challenge/] This dataset is obtained during an ICON project (2017-2018) in collaboration with KU Leuven (ESAT-STADIUS), UZ Leuven, UCB, Byteflies and Pilipili. The goal of this project was to design a system using Behind the ear (bhE) EEG electrodes for monitoring the patient in a home environment. This way, a nice balance can be found between sufficient accuracy of seizure detection algorithms (because EEG is used) and wearability (bhe EEG is relatively subtle, similar to a hear-aid device). The dataset acquired in the hospital during presurgical evaluation. During such presurgical evaluation, neurologists try to see if a specific part of the brain is causing the seizures, and if so, if that part of the brain can be removed during surgery. During the presurgical evaluation, patients are monitored using the vEEG for multiple days (typically a week). Patients are however restricted to move within their room because of the wiring and video analysis. In this dataset, following data is available per patient: • Full 10-20 scalp EEG data of the patient during the presurgical evaluation. • Behind-the-ear data (2 sensors positioned behind each ear) • Single-lead ECG data (typically lead II) Seizures are annotated by the clinicians based on the gold standard vEEG system. These seizure annotations are also available in the dataset. In total 82 patients were recorded between 23/01/2017 and 26/10/2018. From those patients, 54 were recorded with the bhe channels. Forty-two of those patients had seizures during their presurgical evaluation, while for twelve patients no seizure has been recorded. The number of seizures per patient ranged from 1 to 22, with a median of 3 seizures per patient. The duration of the seizures, the time difference of seizure EEG onset and end, varied between 11 and 695 seconds with a median of 50 seconds. 89% of the seizures were Focal Impaired Awareness seizures. 91% of the seizures originated from the (fronto-) temporal lobe. In the folder ’Data’ the raw data in the form of .edf, are provided with annotations for all the patients. The annotations are provided in .tsv (tab separated values) files. For every seizure the first column represents the starting point (in seconds) of the seizure, the second one the end point of the seizure, the third one the type of the seizure, while in the last column extra information are provided. The extra information includes the origin of the seizure, the hemisphere and if the seizure can be noted from the behind the ear channels (bhe:1 in that case). In the header section of every file information concerning the dataset and the annotations used are included. For every subject and for every session (even if no seizure is present) two different sets of annotations are provided. The ”a1”set of annotations is the annotations as provided by the doctors. The ”a2” set of annotations are the annotations used in [2] for training of the algorithm. The annotations provided from the doctors were not always perfectly aligned with the typical rhythmic ictal pattern, hence in ”a2” a refinement of the start of each annotation was performed visually by an engineer. Furthermore, in the annotations of the doctor the end point of some seizures was missing (”none”) in the ”a2” subset of annotations each seizure was considered with a stable length of 10 seconds.