lab-neuro-comp / p2 Goto Github PK

Some parts of the contained documentation are marked as TODO and some others do not reflect the current state of the program. Is it possible to fix it?

It might be a good idea to add test cases to the library documentation, for obvious reasons.

One way to make ICA faster is to resample the EEG signal to 200Hz. This compresses the signal respecting EEG's Nyquist frequency, thus requiring the algorithm to take more steps. Why not?

Some suggestions made during classes:

• Rename "README.md" to "README.txt"
• Add links to documentation and some examples

Currently, the MATLAB path needs to be set as the Protolize root folder so this program can run. Is it possible to make this program run like EEGLab e.g. the protolize command would start it anywhere in the system?

After each processing step, there is a window confirming a step's ending. The problem is that this window does not disappear when "Ok" is clicked.

I can't stop a processing procedure on EEGModule unless I a press ctrl-C at MATLAB terminal. Shouldn't there be a button for that?

Since this is a project under constrution yet used by some people on many different computers, maybe we could work on a mechanism to update the program inside itself; like an update button somewhere or something like that.

When I cancel the search for a folder in the EEG processing module, Protolize deletes the old path and replace it with a zero, probably because that's what the search returned. Can't I keep the old path instead?

Found bug with the limits of the scales when one tries to display just a portion of the whole signal and, afterwards, tries to shift back to the original view.

Pretty self explanatory enhancement suggestion. Also, the parameters used on other modules could appear here as well.

If the user opens MATLAB, runs Protolize!! and tries to use either the EEG Processing module or the EMG Processing module, an error appears in the Command Window: Undefined function or variable "ExcelWorkbook". This happens after the user hits the Run button to start the analysis of the excel file to get the names of the files that'll be processed.

The error disappears once the Process button is hit again. This error also doesn't show show up again not matter how many time you quit Protolize!!, but it pops up again if MATLAB is restarted.

There's a commented line (bad programming habits there tsc tsc) that indicates EEGModule would be able to reject some channels, but the script does not work when this line is uncommented. Why??

This problem happens while using the stft module when it tries to create a newsignal array using the signal_with_padding function.

I had some problems testing the lowpass filter on the Time/Spec module. The test signal fc5 wasn't filtered using the LPF on 75Hz. Is this filter always working?

Based on the works done on eegmodule, it is possible to get rid of the global variable at voicemodule that is being used to transfer data from one window to another. In summary, it will consist of turning a window into a function and then call it from the main script.

The newer versions of EEGLab raise some errors when using the EEG module on Protolize!, specially when it comes to running ICA and locating channels (the application performs the mapping but the locations do not appear). Why is it so?

In EEG module, there are some hard coded numbers, referring to channels that may be removed and to the reference channel. Why not turning these numbers into parameters to be asked for before running the module's script?

The jar for dealing with EDF files could be part of the P2Lib, as well as a function for converting EDF to ASCII files. This will create another feature for Protolize, acessible for outside scripts.

I want a checkmark list with only one checkmarked item on a toolbar menu on GUIDE. How can I do this?