Quantification of stroke lesion volume using epidural EEG in a cerebral ischaemic rat model

We have uploaded the dataset of rat EEG in response to the specific sound stimuli. The dataset includes rat EEG of normal subjects (n = 10), mild (n = 7), moderate (n = 7), and severe (n = 7) right auditory cortical infarction subjects. 

The dataset can be analyzed using MATLAB software. For power spectrum density (PSD) analysis of the data, we used the zero-phase forward and reverse Infinite Impulse Response Butterworth filter of 4th order. Further, we averaged the last 300 ms of the signal before stimulus onset as a baseline correction. Next, we down-sampled the data from 1,200 to 600 Hz. Subsequently, we conducted the PSD analysis using signals obtained from the target stimulus onset to 1 s using Welch’s method, which is one of the most widely used periodogram methods for determining the power density of EEG frequency components. The parameter was set to divide the EEG signals into eight sections of equal length, each with a 50% overlap based on the Hamming window. We defined the frequency range for each band as follows: Delta (1-4 Hz), theta (4-8 Hz), alpha (8-12 Hz), and beta (12-30 Hz). We calculated the relative power for each frequency band by summing all of the absolute PSD values across the four bands to compute the total power followed by dividing the absolute value for each frequency band with the total power. Finally, we calculated the DAR (delta/alpha ratio) and the DTABR ((delta + theta) / (alpha + beta) ratio), which were computed by the relative power of the relevant frequency bands. Further, we analysed AEPs in response to the target sound stimuli by averaging all of the epochs between 300 ms before the stimuli onset to 500 ms after it. The AEP amplitude was defined as the highest recorded voltage following the sound stimulus. The latency of the components of the AEPs was defined as the duration from stimulus onset to the peak amplitude.