Dataset of quantitative spectral EEG of different stages of kindling acquisition in rats

The data represented here are in relation with the manuscript "Quantitative assessments of extracellular EEG to classify specific features of main phases of seizure acquisition based on kindling model in Rat" (Jalilifar et al., 2017) [1] which quantitatively classified different main stages of the kindling process based on their electrophysiological characteristics using EEG signal processing. The data in the graphical form reported the contribution of different sub bands of EEG in different stages of kindling- induced epileptogenesis. Only EEG signals related to stages 1–2 (initial seizure stages (ISSs)), 3 (localized seizure stage (LSS)), and 4–5 (generalized seizure stages (GSSs) were transferred into frequency function by Fast Fourier Transform (FFT) and their power spectrum and power of each sub bands including delta (1–4 Hz), Theta (4–8 Hz), alpha (8–12 Hz), beta (12–28 Hz), gamma (28–40 Hz) were calculated with MATLAB 2013b. Accordingly, all results were obtained quantitatively which can contribute to reduce the errors in the behavioral assessments.


a b s t r a c t
The data represented here are in relation with the manuscript "Quantitative assessments of extracellular EEG to classify specific features of main phases of seizure acquisition based on kindling model in Rat" (Jalilifar et al., 2017) [1] which quantitatively classified different main stages of the kindling process based on their electrophysiological characteristics using EEG signal processing. The data in the graphical form reported the contribution of different sub bands of EEG in different stages of kindling-induced epileptogenesis. Only EEG signals related to stages 1-2 (initial seizure stages (ISSs)), 3 (localized seizure stage (LSS)), and 4-5 (generalized seizure stages (GSSs) were transferred into frequency function by Fast Fourier Transform (FFT) and their power spectrum and power of each sub bands including delta (1-4 Hz), Theta (4-8 Hz), alpha (8-12 Hz), beta (12-28 Hz), gamma (28-40 Hz) were calculated with MATLAB 2013b. Accordingly, all results were obtained quantitatively which can contribute to reduce the errors in the behavioral assessments.  Data accessibility All of the data presented in this study are accessible within this article

Value of the data
The data show the differences between the EEG signals of the kindling and control animals.
The reported data can be used to develop seizure prediction model for temporal lobe epilepsy using EEG signals.
Our data can contribute to explore the patterns of the kindling-induced epileptogenesis progression which can be useful to develop antiepileptic approach.

Data
The data of this study were collected from an animal in vivo study aiming at quantitative assessment of epileptogenesis in a rapid kindling model in rats [1]. Considering the unique features of EEG for seizure prediction [2], these data present the raw data of spectral analyses of the field potentials recorded during the progression of Amygdala kindling in rats to determine the quantitative features of main phases of kindling acquisition. In this paper, stages 1 and 2 of kindling were considered initial seizure stages (ISSs), stage 3 as localized seizure stage (LSS), and stages 4 and 5 as generalized seizure stages (GSSs). Tables 1-3 present the spectral powers of different sub bands of EEGs in ISSs, LSSs, and GSSs of the kindling process, respectively. Moreover, Table 4 presents percentage of different sub bands power in the control group.

Materials and methods
Adult male rats weighing 200 710 g were housed individually under standard conditions (an ambient temperature (25 72°C) and 12-h light: 12-h dark: 12-h light cycle).
Rats were randomly divided into two groups (ten for the kindle group and 6 for sham) and anesthetized under intraperitoneal injection of ketamine (100 mg/kg) and Xylazine (10 mg/kg) mixture [3]. One tripolar stainless steel electrode (a bipolar for stimulating and a monopole for recording EEG signal) was implanted in amygdala using Paxinos and Waston atlas coordinates: for amygdala targeting, anteroposterior: -2.5 mm; lateral: 4.8 mm; vertical: 7.2 and 0.2 mm below the skull [4]. Three holes were drilled, one for positioning a monopolar electrode attached to a screw which was located near the frontal lobe as ground and reference, the two for anchor screws. Electrodes and screws were fixed using acrylic dental cement and attached to a socket. The protocol of Table 2 The percentage of different frequencies in LSSs. We reported the mean value for each rat.

Rats
Delta Theta Alpha Beta Gamma  Table 4 Contribution of different sub bands power in sham group. We reported the mean value for each rat. . Following a 10-day recovery period after surgery, the threshold intensity was determined using a 3 s of monophasic square wave of 50 Hz initially applied at 30 µA and it was increased in step of 15 µA at 15 min intervals until emerging at least 6 s of afterdischarges (ADs). All rats in the kindle group were subjected to daily stimulation using a 3 s train of 50 Hz monophasic pulses of 1ms duration with threshold intensity which were applied 12 times daily with 5 min intervals [5], whereas sham animals only experienced stimulation condition and received placebo stimulation (Fig. 1). Therefore, the EEG of sham animals can be considered as a baseline. Behavioral development of kindling acquisition was scored according to Racine stages [6]. This process was continued until emerging stage 5 of kindling. EEG signals recorded from the implanted electrode in the amygdala and monitored with electro module system (Tehran, Iran) which was connected to computer using e-probe software. During kindling acquisition, we could save the starting and ending time of each stage of kindling as a text file (an event file) which can be considered in extracting each stage. Data were digitized at a sampling rate of 10 KHz. Moreover, the electro module automatically applied a filter on 50 Hz frequency to remove DC effect from the signals. Recorded EEG signals were saved as binary files. These binary files were then imported into EEGLAB software for pre-processing stage. Moreover, a band-pass filter between 0.5-60 was applied to remove the effect of other frequencies. In the EEGLAB, we separated the EEG signals of each stage and the obtained signals were saved as dataset files which can be imported into MATLAB. These signals were then transferred into frequency domain by Fast Fourier Transform (FFT) and MATLAB 2013b was used to calculate their power spectrum and power of each sub bands including delta (1-4 Hz), Theta (4-8 Hz), alpha (8-12 Hz), beta (12-28 Hz), and gamma (28-40 Hz).