Hello!
I am currently building a new pipeline for EEG data analysis within EEGLab, and in the past I've had undergraduate research assistants do a first pass at visually cleaning the data to remove muscle movement and large artifacts, followed by a quick check by me or other senior EEG folks before moving on. I would like to implement something similar within EEGLab so I can double check my undergrads' work before actually removing blocks of clearly messy data and moving onto ICA artifact correction and epoching.
I see that bad chunks of continuous data are removed within EEGLab and boundary events are placed in the file, but I cannot see how to save a file that has bad data marked WITHOUT deleting those marked blocks. Not being able to see what is marked as bad make it nearly impossible for me to check my undergrads' work, as I cannot see what was deleted. I've tried opening the continuous data with Tools > "Reject continuous data by eye" and Plot > "Channel data (continuous scroll)" followed by selecting either "REJECT" or Figure > "Accept and close" (both of these options delete the data and insert boundary events), and if I click CANCEL or Figure > "Cancel and close" no sections that were marked are saved.
Any help on how to save marked bad sections of continuous data WITHOUT removing them within EEGLab or with a plugin would be much appreciated! Unfortunately I am in a new lab, and so all I have access to is EEGLab/ERPLab, or any other free/open-source software.
Thanks!
Hello,
Yes it is possible to do this. Your best bet is to write a code for Matlab/EEGlab where your students have to manually enter the trials or times they want to reject and instruct them not to push the button that rejects trials on the GUI. Then you go in and check if the periods they've manually entered are things you agree with. If you want more detailed instructions, feel free to email me.
Many thanks to everyone who helped me piece this together! Below is the answer I finally came up with, hope it is helpful to others.
First, go through the continuous window and reject bad sections. Click REJECT, but then cancel for saving a new file. That creates a TMPREJ matrix with the info needed. Then:
badSection = TMPREJ(:,1:2);
%saves the matrix with reject info as new matlab variable clear TMPREJ;
%clears matlab variable used by EEGLab save('Z:\YourPathHere\badSectionSN001.mat','badSection');
%saves new matlab variable for your particpant clear badSection.
%clears the matlab variable for you to repeat on next subject
To check each subject, load the participant's .set. Then:
clear badSection;
%makes sure this variable is cleared
load('Z:\YourPathHere\badSectionSN001.mat');
%loads the badSection matrix for subject number 1 eegplot(EEG.data, 'winrej', badSection, 'srate', 500, 'events', EEG.event);
%plots the EEG data loaded with bad sections added. 'srate' is sampling rate (updated as needed to get seconds to line up, default is 256) and 'events' is to plot the event markers. Can also change how many seconds are displayed with 'winlength' followed by a number for seconds displayed. See eegplot help for other useful commands.
Then you can scroll through and adjust rejections as needed. Then go to Figure-> Accept and Close which doesn't perform rejections but creates a new TMPREJ variable with the updated rejections. Then repeat the first step with the approved rejected sections:
badSectionFinal = TMPREJ(:,1:2);
%saves the matrix with reject info as new matlab variable clear TMPREJ;
%clears matlab variable used by EEGLab save('Z:\YourPathHere\badSectionFinalSN001.mat','badSection'Final);
%saves new matlab variable for your participant clear badSectionFinal.
%clears the matlab variable for you to repeat on next subject Then finally, apply the rejections! Load the participant's .set, then: load('Z:\YourPathHere\badSectionFinalSN001.mat') EEG = eeg_eegrej( EEG, badSectionFinal);
- Badges
- Science topic
- Similar topics
- Filing