Intracranial EEG (iEEG) Recording in a Rat Model of Epilepsy

Citation Author(s):
Azam
Ghanaei
Iran University of Science and Technology
Abbas
Erfanian
Iran University of Science and Technology
Submitted by:
Azam Ghanaei
Last updated:
Wed, 03/05/2025 - 03:52
DOI:
10.21227/vka0-hn24
Data Format:
**.mat
License:
Creative Commons Attribution
192 Views
Categories:
Signal Processing
Biomedical and Health Sciences
Biophysiological Signals
Keywords:
epileptic seizure, convolutional neural network, recurrence plot, spike-wave discharge (SWD)
0
0 ratings - Please login to submit your rating.
ACCESS DATASET CITE SHARE/EMBED

Abstract 

This database, collected at the Neural Engineering Laboratory, Iran University of Science and Technology, comprises iEEG recordings from Wistar rats during healthy and epileptic conditions. Recordings were collected from 5 rats (3 males, 2 females, weighing 260-378 g and aged 4-5 months). iEEG signals were recorded from 3 brain sites: motor cortex (left M1), thalamus (left ANT), and hippocampus (right CA1) of freely moving rats. As a result, for each rat, a matrix with 3 columns (representing the 3 signals) is available in this dataset.

To induce temporal lobe epilepsy (TLE), a nonconvulsive pentylenetetrazol (PTZ) model was used. In this epileptic model, rats displayed recurrent spontaneous seizures at Racin stage 1-2, accompanied by spike-wave discharges (SWDs). A seizure can be detected by identifying high-amplitude, repetitive SWDs in the epileptic signals.

Recordings were conducted in two conditions: the healthy condition (before PTZ injection) and the epileptic condition (after PTZ injection). Each healthy and epileptic condition contains several trials.

The details of recording and experimental methods can be seen in the paper:

Azam Ghanaei and Abbas Erfanian, ''Real-time Seizure Prediction Using Nonlinear Dynamical Analysis of Intracranial EEG and Deep Learning in a Rodent Model of Epilepsy,'' Submitted to IEEE trans. Biomed. Eng. 2025.

Instructions: 

Data files organization:

The data files are organized into five main folders, each corresponding to a subject (Rat01, Rat02, Rat03, Rat04, and Rat05). Each main folder contains two subfolders for the healthy and epileptic conditions, named 'Healthy' and 'Epileptic'. Each subfolder contains several .mat files corresponding to the recording trials. Each .mat file contains three iEEG signals from a single recording trial. 

The name of each .mat file is 'RatID#_E/H_S#,' where 'ID#' refers to the rat’s number, 'E/H' indicates the animal conditions (E for 'Epileptic' and H for 'Healthy'), and 'S#' indicates the experimental trial. For example, 'Rat01_E_01.mat' contains the recorded iEEG signals from the epileptic animal for rat 1, trial 1.

Data format:

The data are in MATLAB format (.mat). Each .mat file contains a signal matrix (i.e., 'RatID#_E/H_S#') with n rows and mcolumns (), representing iEEG signals from a single trial.

  • n is the number of samples, which can be calculated as Sampling Rate × Trial Recording Time.
  • m is the number of channels (m=3)

To have signals in the millivolt range, divide the amplitude by 10000. 

 Data information:

*********************************** Rat01 ***********************************

  • Gender: Male
  • Weight: 370 g
  • Age: 5 months
  • Data Sampling Rate: 6024 Hz
  • Channels (as rows of signal matrix 'RatID#_E/H_S#'):
    • Row 1: CA1 channel
    • Row 2: ANT channel
    • Row 3: M1 channel
  • Session Recording Time: 150 second
  • Number of trials in healthy file: 2
  • Number of trials in epileptic file: 6

 

*********************************** Rat02 ***********************************

  • Gender: Female
  • Weight: 260 g
  • Age: 4 months
  • Data Sampling Rate: 6000 Hz
  • Channels (as rows of signal matrix 'RatID#_E/H_S#'):
    • Row 1: CA1 channel
    • Row 2: ANT channel
    • Row 3: M1 channel
  • Session Recording Time: 150 second
  • Number of trials in healthy file: 1
  • Number of trials in epileptic file: 6

 

*********************************** Rat03 ***********************************

  • Gender: Female
  • Weight: 267 g
  • Age: 4 months
  • Data Sampling Rate: 6000 Hz
  • Channels (as rows of signal matrix 'RatID#_E/H_S#'):
    • Row 1: ANT channel
    • Row 2: CA1 channel
    • Row 3: M1 channel
  • Session Recording Time: 140 second
  • Number of trials in healthy file: 1
  • Number of trials in epileptic file: 13

 

*********************************** Rat04 ***********************************

  • Gender: Male
  • Weight: 378 g
  • Age: 5 months
  • Data Sampling Rate: 6000 Hz
  • Channels (as rows of signal matrix 'RatID#_E/H_S#'):
    • Row 1: CA1 channel
    • Row 2: ANT channel
    • Row 3: M1 channel
  • Session Recording Time: 180 second
  • Number of trials in healthy file: 1
  • Number of trials in epileptic file: 8

 

*********************************** Rat05 ***********************************

  • Gender: Male
  • Weight: 315 g
  • Age: 5 months
  • Data Sampling Rate: 6000 Hz
  • Channels (as rows of signal matrix 'RatID#_E/H_S#'):
    • Row 1: CA1 channel
    • Row 2: ANT channel
    • Row 3: M1 channel
  • Session Recording Time: 180 second
  • Number of trials in healthy file: 2
  • Number of trials in epileptic file: 8

 

 

Dataset Files

LOGIN TO ACCESS DATASET FILES

Documentation

AttachmentSize
File Data-Description.pdf432.86 KB

 

QUESTIONS?