Identified important EEG channels using recurrent analysis in patients with chronic migraine

This database contains calculated indicators of channel significance based on the application of recurrent analysis to EEG data using the example of the article Runnova A, Selskii A, Emelyanova E, Zhuravlev M, Popova M, Kiselev A, Shamionov R. Modification of Joint Recurrence Quantification Analysis ( JRQA) for assessing individual characteristics from short EEG time series. Chaos. 2021;31(9):093116. doi: 10.1063/5.0055550. For each EEG channel of each subject, the channel significance indicator was separately calculated for all presentations of stimuli in the cognitive test, which are located in the SR folder. The RR_Simple_Hard folder contains the results of calculating significant channels separately for simple presentations - 3 and 4 squares, and complex presentations - 7 and 8 squares. A description of the EEG setup, subjects, and experimental design is provided below.
Test subjects volunteered to participate in the experiment. The biomedical data were processed honoring the confidentiality and anonymity of the study respondents. All procedures in studies involving human participants were conducted in accordance with the principles of the Declaration of Helsinki and approved by the Research Ethics Committee at Saratov State Medical University. Written informed consent was given by all participants.  The participants were diagnosed in compliance with the diagnostic criteria of the International Classification of Headache Disorders (ICHD-3, beta version). We used the migraine questionnaire to evaluate the data on migraine attacks and disease courses. The exclusion criteria were acute headache, other neurological disorders, and intake of medicines targeting the central nervous system (including medications for migraine prophylaxis) within the preceding 24 h, as well as a Beck Depression Inventory (BDI) score >12 points. The  sample comprised 23 subjects.
The study group included patients with chronic and recurrent migraine. The subjects had no other serious health problems, or drug or alcohol addictions. Patients were 27–66 years of age.
During all experiments, the multichannel EEG data were recorded using the EEG recorder Encephalan-EEGR-19/26 (Medicom MTD, Taganrog, Russian Federation). The data were recorded at a sampling rate of 500 Hz using the conventional monopolar recording technique with two reference electrodes and multiple recording electrodes (n = 31). EEG signals were obtained using special headcaps with prewired Ag/AgCl adhesive electrodes. Two reference electrodes, A1 and A2, were located on the mastoid processes, and the ground electrode (N) was placed above the forehead. EEG signals were filtered using a bandpass filter with cutoff frequencies of 0.5 Hz (high pass filter) and 30 Hz (low pass filter), and a notch filter of 50 Hz. The experiments were carried out in the early afternoon hours at a specially equipped laboratory. During monitoring, the subjects were in a comfortable reclining position in a neurophysiological chair with support for the neck and the back of the head in order to avoid the occurrence of artifacts associated with muscle tension in these areas. The neuropsychological experiment allowed for distinguishing cognitive ERPs. At the beginning and end of the experiment, passive wakefulness was recorded for 10 min each time. In between, the active part of the experiment was recorded for ~40 min. During the active part, short presentations of visual stimuli were delivered in a pseudorandom sequence. For all subjects, the order of stimulus presentation, duration of stimulus presentation, and duration of pauses between stimuli were identical. Each subject observed 350 stimuli. The stimulus was presented for 0.4–0.8 s. After its presentation, during a pause until the next stimulus, the subject saw solely a gray screen. Each visual stimulus per se was an image with a different number of squares (3–8). The task of the subject was to estimate an even or odd number of squares and make a choice by pressing one of two buttons on the remote control.