eegAnalysis (version 0.0)

plotEEG: Plot EEG data

Description

This function was designed to do different types of plots of EEG data. Graphs of the original data, of the spectrum, continuous wavelet transform and t-value scalogram of the signals can be plotted. The main idea is to help the user to find nice features to use in his final model.

Usage

plotEEG(data, classes.Id, rec.Id, which.classes = "ALL", which.rec = "ALL", 
which.channels = "ALL", type = "original", m.a = 1, n.colors = 200, 
wavelet = "gaussian2", abs = FALSE, Real = TRUE, variance = 1)

Arguments

data

is the data frame containing the EEG signals. The data frame must be organized as follows: each column represents a different channel. Thus, the signals collected by each channel are represented in each row. To identify the class and recording represented by each line, the vectors classes.Id and rec.Id should be given and must have length equal to the number of rows in the data frame.

classes.Id

is a vector with length equal to the number of rows in data. Thus each value in the array identifies the class ID of each signal in each row of the database. For example, let classes.Id = c(rep(1,5),rep(2,5)), this means that the first 5 rows of data represents the class with ID 1 and the lines 6 to 10 represent the class with ID 2.

rec.Id

is a vector with length equal to the number of rows in data. Thus each value in the array identifies the recording ID of each signal in each row of the database. For example, let rec.Id <- c(rep(1,5),rep(2,5)), this means that the first 5 rows of data represents the recording with ID 1 of some class and the lines 6 to 10 represent the recording with ID 2 of some class. rec.Id must be numeric and numerated from 1 to the total number of recordings for each class.

which.classes

a vector representing which classes will be plotted. For example, if which.classes= c(1,3) then the classes with ID 1 and 3 will be plotted (indicated by the vector classes.Id). If which.classes= "ALL" then all classes will be plotted. Obs: if type = "T.pvalue" then only two classes are allowed.

which.rec

a list representing which recordings shall be plotted for each class. For example, if which.rec= list(c(1,3,4),c(1,2,4)) then the recordings with ID 1,3 and 4 will be plotted for the first class indicated in the vector which.classes and the recordings with ID 1,2 and 4 will be plotted for the second class indicated in the vector which.classes. If which.rec="ALL" then all recordings will be plotted.

which.channels

a vector representing which channels shall be plotted. For example, if which.channels =c(1,3) then the channels 1 and 3 (columns 1 and 3 of data) will be plotted. If which.channels="ALL" then all channels will be plotted.

type

if type="original" then the original recordings will be plotted, if type= "spectrum" then the spectrum of each recording will be plotted, if type="wavelet" then the continuous wavelet transform matrix of the recordings will be plotted, if type="T.pvalue" then the t-value scalogram of the recordings will be plotted (see Bostanov 2004).

m.a

is the moving average parameter, if m.a=1 it is not used. If type= "original" then the moving average is applied for each channel of the original recordings. If type= "spectrum" and m.a is a number, the moving average is applied on the spectrum of the recordings, otherwise, if m.a is a vector of size 2, the first value defines the moving average for the original recordings and the second value defines the moving average for the spectrum. If type= "wavelet" or "T.pvalue", m.a should be a vector of size 2, indicating the dimensions in which the moving average will be applied on the CWT matrix.

n.colors

is the number of colors for the contour plot. Used if type="wavelet" or type= "T.pvalue".

wavelet

if type="wavelet" defines which wavelet is used. See wavCWT for more details.

abs

if TRUE then the absolute value of the continuous wavelet transform matrix is used. Used if type="wavelet" or type="T.pvalue".

Real

if TRUE takes the real part of the continuous wavelet transform matrix, if FALSE takes the imaginary part. Used if wavelet="morlet".

variance

the variance parameter for the continuous wavelet transform. See wavCWT for more details.

References

Bostanov, V. (2004) BCI Competition 2003 - Data Sets Ib and IIb: Feature Extraction From Event-Related Brain Potentials With the Continuous Wavelet Transform and the t-Value Scalogram. IEEE transactions on biomedical engineering, V. 51, no. 6.

Hastie, T., Tibshirani, R., Friedman, J. (2009) The Elements of Statistical Learning: Data Mining, Inference, and Prediction. 2nd ed. Stanford: Springer.

Brockwell, P.J., Davis, R.A. (2002) Introduction to Time Series and Forecasting. 2nd ed. Colorado: Springer. Cap. 4.

See Also

wavCWT, spectrum

Examples

library(eegAnalysis)

#### Simulating some data
sim<-randEEG()

#### Plot some of the recordings:
plotEEG(sim$data, sim$classes.Id, sim$rec.Id, which.classes = "ALL", 
        which.rec=list(c(1,2,3),c(2,3,4)), which.channels=c(1,2,3), 
        type = 'original', m.a = 10) 


#### Plot the spectrum
plotEEG(sim$data, sim$classes.Id, sim$rec.Id, which.classes = "ALL", 
        which.rec="ALL", which.channels=1, type = 'spectrum', m.a = 10) 


#### Plot the spectrum 
plotEEG(sim$data, sim$classes.Id, sim$rec.Id, which.classes = "ALL", 
        which.rec="ALL", which.channels=5, type = 'spectrum', m.a = c(5,20)) 


#### Plot the continuous wavelet transform
plotEEG(sim$data, sim$classes.Id, sim$rec.Id, which.classes = 1, 
        which.rec=list(c(1)), which.channels=c(2), type = 'wavelet',
        wavelet="gaussian2", abs=TRUE,variance=1) 


### Plot the T-value scalogram
plotEEG(sim$data, sim$classes.Id, sim$rec.Id, which.classes = "ALL", 
        which.rec="ALL", which.channels=c(2), type = 'T.pvalue', 
        wavelet="gaussian2", abs=TRUE,variance=10)