Skip to content

Commit

Permalink
Correct similarity computation Bug, update of acknowledgments and funds
Browse files Browse the repository at this point in the history 
Corrected similarity computation bug,
updated acknowledgment with inclusion of BIREHAB EU H2020 grant
  • Loading branch information
@flower1345
flower1345 committed Oct 18, 2019
1 parent ce3d4dd commit bdc39e8
Show file tree
Hide file tree
Showing 7 changed files with 118 additions and 40 deletions.
13 changes: 9 additions & 4 deletions README.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -10,7 +10,7 @@ Next, for each IC in each decomposition RELICA computes the dipolarity of the IC
All plug-ins in EEGLAB, including RELICA, can be installed in two ways. To install RELICA:

1. **From the EEGLAB Plug-in Manager:** Launch EEGLAB and select menu item **File > Manage EEGLAB Extensions** in the main EEGLAB window. A plug-in manager window will pop up. Look for and select the RELICA plug-in, then press **Install/Update**.

2. **From the web:** Download the RELICA plug-in zip file either from [this](https://github.com/sccn/relica) GitHub page (select ‘Download Zip‘) or from [this EEGLAB wiki plug-ins page](https://sccn.ucsd.edu/wiki/Plugin_list_all) (select **RELICA**). Decompress the zip file in the plug-ins folder in the main eeglab folder (*../eeglab/plugins/*).

Restart EEGLAB. If the installation is successful, a menu item for calling RELICA will appear in the EEGLAB menu at **Tools > Run RELICA** .
Expand Down Expand Up @@ -79,7 +79,7 @@ In the figure above, the black dots are individual bootstrap ICs, and red intege

#### Plot type 2: Cluster exemplar maps
Select the option **Cluster exemplar maps** from the **Plot type** menu in the *pop\_relica_results* window, then press **Plot**. A figure like the one below will be displayed.

<img src="./images/relica_realicmaps.jpg" width="800">

Above, the title of each plot panel shows the cluster number and the component stability index from [Himberg et al. 2004](https://www.sciencedirect.com/science/article/pii/S1053811904001661) (heuristic measure of the quality of the estimates, ideally 100%) for all IC clusters.
Expand All @@ -88,10 +88,15 @@ Above, the title of each plot panel shows the cluster number and the component s
To plot the scalp maps of the ICs in a single IC cluster, select option **Cluster scalp maps** and then specify the cluster number to plot (**Cluster to plot**), using the cluster numbers shown in the **Bootstrap IC clusters** plot (type 5 above, corresponding to IC 3). Specify the maximum number of IC scalp maps from the cluster to plot (**Max number of maps**). Pressing **Plot** displays the figure below.

<img src="./images/relica_cluster1.jpg" width="400">

Above, each figure panel shows the scalp map for a member of the selected IC cluster (default: 1), beginning with the first bootstrap. When the number of cluster ICs (and bootstraps) is large, only the requested number are shown (default: one page of 20 maps).

## Running and visualizing RELICA from MATLAB command line
Running RELICA, either locally or using NSG, and retrieving and visualizing the results can also be performed from the MATLAB command line or by a MATLAB script. For syntax of these calls, refer to the help messages for the functions *pop\_relica.m* and *pop\_relica_results.m*.


## References

1) Artoni, F., Menicucci, D., Delorme, A., Makeig, S., & Micera, S. (2014). *RELICA: a method for estimating the reliability of independent components*. NeuroImage, 103, 391-400.

2) Artoni, F., Delorme A., Makeig S. (2018) *Applying dimension reduction to EEG data by Principal Component Analysis reduces the quality of its subsequent Independent Component decomposition*, Neuroimage 175 176-187

22 changes: 15 additions & 7 deletions eegplugin_RELICA.m
View file
Original file line number Diff line number Diff line change
Expand Up @@ -8,18 +8,26 @@
% trystrs - [struct] "try" strings for menu callbacks.
% catchstrs - [struct] "catch" strings for menu callbacks.
%
% Author: Fiorenzo Artoni, The Biorobotics Institute / EPFL, 2017 %
%
% Reference:
% Author: Dr. Fiorenzo Artoni EPFL, 2019 %
%
% References:
% (1) Artoni, F., Menicucci, D., Delorme, A., Makeig, S., & Micera, S. (2014).
% RELICA: a method for estimating the reliability of independent components.
% NeuroImage, 103, 391-400.
% NeuroImage, 103, 391-400.
%


% Copyright (C) 2017 Fiorenzo Artoni, The Biorobotics Institute , EPFL, SCCN, [email protected]
% (2) Artoni, F., Delorme A., Makeig S. (2018)
% Applying dimension reduction to EEG data by Principal Component Analysis
% reduces the quality of its subsequent Independent Component
% decomposition, Neuroimage 175 176-187
%
% This project was in part supported by the European Union's Horizon 2020
% research and innovation programme under Marie Sklodowska-Curie Action
% agreement no. 750947 (BIREHAB)
%
% Acknowledgments go to Ramon Martinez-Cancino (SCCN/INC/UCSD 2019) for making the
% algorithm available and parallelized on the NSG server and including other ICA algorithms
% and Arnaud Delorme and Scott Makeig (SCCN/INC/UCSD 2019) for the constant inputs
% and ideas to perfect the functionality.
% Clustering and relative visualization within RELICA makes use of modified
% routines from J. Himberg's open source FastICA - ICASSO package
% Beamica is part of C. Kothe's open source BCILAB toolbox
Expand Down
27 changes: 23 additions & 4 deletions pop_relica.m
View file
Original file line number Diff line number Diff line change
Expand Up @@ -71,10 +71,29 @@
%
% See also: relica.m
%
% Authors: Fiorenzo Artoni, The Biorobotics Institute / EPFL, 2019
% Ramon Martinez-Cancino SCCN/INC/UCSD 2019

% Copyright (C) 2019 Ramon Martinez-Cancino and Fiorenzo Artoni
% Author: Dr. Fiorenzo Artoni, 2019 %
%
% References:
% (1) Artoni, F., Menicucci, D., Delorme, A., Makeig, S., & Micera, S. (2014).
% RELICA: a method for estimating the reliability of independent components.
% NeuroImage, 103, 391-400.
%
% (2) Artoni, F., Delorme A., Makeig S. (2018)
% Applying dimension reduction to EEG data by Principal Component Analysis
% reduces the quality of its subsequent Independent Component
% decomposition, Neuroimage 175 176-187
%
% This project was in part supported by the European Union's Horizon 2020
% research and innovation programme under Marie Sklodowska-Curie Action
% agreement no. 750947 (BIREHAB)
%
% Acknowledgments go to Ramon Martinez-Cancino (SCCN/INC/UCSD 2019) for making the
% algorithm available and parallelized on the NSG server.
% Acknowledgments go also to Arnaud Delorme and Scott Makeig (SCCN/INC/UCSD 2019)
% for the precious inputs and ideas to perfect the project.
% Clustering and relative visualization within RELICA makes use of modified
% routines from J. Himberg's open source FastICA - ICASSO package
% Beamica is part of C. Kothe's open source BCILAB toolbox
%
% This program is free software; you can redistribute it and/or modify
% it under the terms of the GNU General Public License as published by
Expand Down
20 changes: 15 additions & 5 deletions pop_relica_load.m
View file
Original file line number Diff line number Diff line change
Expand Up @@ -11,16 +11,26 @@
% EEG - Input dataset
% file - Complete path of RELICA.mat file
%
% Author: Fiorenzo Artoni, The Biorobotics Institute / EPFL, 2017 %
% Author: Dr. Fiorenzo Artoni, 2019 %
%
% References:
%
% (1) Artoni, F., Menicucci, D., Delorme, A., Makeig, S., & Micera, S. (2014).
% RELICA: a method for estimating the reliability of independent components.
% NeuroImage, 103, 391-400.

% Copyright (C) 2017 Fiorenzo Artoni, The Biorobotics Institute , EPFL, SCCN, [email protected]
% NeuroImage, 103, 391-400.
%
% (2) Artoni, F., Delorme A., Makeig S. (2018)
% Applying dimension reduction to EEG data by Principal Component Analysis
% reduces the quality of its subsequent Independent Component
% decomposition, Neuroimage 175 176-187
%
% This project was in part supported by the European Union's Horizon 2020
% research and innovation programme under Marie Sklodowska-Curie Action
% agreement no. 750947 (BIREHAB)
%
% Acknowledgments go to Ramon Martinez-Cancino (SCCN/INC/UCSD 2019) for making the
% algorithm available and parallelized on the NSG server.
% Acknowledgments go also to Arnaud Delorme and Scott Makeig (SCCN/INC/UCSD 2019)
% for the precious inputs and ideas to perfect the project.
% Clustering and relative visualization within RELICA makes use of modified
% routines from J. Himberg's open source FastICA - ICASSO package
% Beamica is part of C. Kothe's open source BCILAB toolbox
Expand Down
25 changes: 17 additions & 8 deletions pop_relica_plots.m
View file
Original file line number Diff line number Diff line change
Expand Up @@ -25,20 +25,29 @@
% Outputs:
% EEG - Output dataset: RELICA data is in EEG.etc.RELICA
%
% Author: Fiorenzo Artoni, The Biorobotics Institute / EPFL, 2017 %
% Ramon Martinez-Cancino, SCCN, INC, UCSD

% Reference:
% Author: Dr. Fiorenzo Artoni, 2019 %
%
% References:
% (1) Artoni, F., Menicucci, D., Delorme, A., Makeig, S., & Micera, S. (2014).
% RELICA: a method for estimating the reliability of independent components.
% NeuroImage, 103, 391-400.
% NeuroImage, 103, 391-400.
%

% Copyright (C) 2017 Fiorenzo Artoni, The Biorobotics Institute , EPFL, SCCN, [email protected]
% (2) Artoni, F., Delorme A., Makeig S. (2018)
% Applying dimension reduction to EEG data by Principal Component Analysis
% reduces the quality of its subsequent Independent Component
% decomposition, Neuroimage 175 176-187
%
% This project was in part supported by the European Union's Horizon 2020
% research and innovation programme under Marie Sklodowska-Curie Action
% agreement no. 750947 (BIREHAB)
%
% Acknowledgments go to Ramon Martinez-Cancino (SCCN/INC/UCSD 2019) for making the
% algorithm available and parallelized on the NSG server.
% Acknowledgments go also to Arnaud Delorme and Scott Makeig (SCCN/INC/UCSD 2019)
% for the precious inputs and ideas to perfect the project.
% Clustering and relative visualization within RELICA makes use of modified
% routines from J. Himberg's open source FastICA - ICASSO package
% Beamica is part of C. Kothe's open source BCILAB toolbox
% Beamica is part of C. Kothe's open source BCILAB toolbox
%
% This program is free software; you can redistribute it and/or modify
% it under the terms of the GNU General Public License as published by
Expand Down
27 changes: 22 additions & 5 deletions relica.m
View file
Original file line number Diff line number Diff line change
Expand Up @@ -50,16 +50,31 @@
% EEG - Output dataset: RELICA data is in EEG.etc.RELICA, the same is
% saved in folder_relica folder
%
% Author: Fiorenzo Artoni, The Biorobotics Institute / EPFL, 2019 %
% Ramon Martinez-Cancino SCCN/INC/UCSD 2019
% Author: Dr. Fiorenzo Artoni, 2019 %
%

% Copyright (C) 2019 Fiorenzo Artoni and Ramon Martinez-Cancino
% References:
% (1) Artoni, F., Menicucci, D., Delorme, A., Makeig, S., & Micera, S. (2014).
% RELICA: a method for estimating the reliability of independent components.
% NeuroImage, 103, 391-400.
%
% (2) Artoni, F., Delorme A., Makeig S. (2018)
% Applying dimension reduction to EEG data by Principal Component Analysis
% reduces the quality of its subsequent Independent Component
% decomposition, Neuroimage 175 176-187
%
% This project was in part supported by the European Union's Horizon 2020
% research and innovation programme under Marie Sklodowska-Curie Action
% agreement no. 750947 (BIREHAB)
%
% Acknowledgments go to Ramon Martinez-Cancino (SCCN/INC/UCSD 2019) for making the
% algorithm available and parallelized on the NSG server.
% Acknowledgments go also to Arnaud Delorme and Scott Makeig (SCCN/INC/UCSD 2019)
% for the precious inputs and ideas to perfect the project.
% Clustering and relative visualization within RELICA makes use of modified
% routines from J. Himberg's open source FastICA - ICASSO package
% Beamica is part of C. Kothe's open source BCILAB toolbox
%
%
% This program is free software; you can redistribute it and/or modify
% it under the terms of the GNU General Public License as published by
% the Free Software Foundation; either version 2 of the License, or
Expand Down Expand Up @@ -278,7 +293,9 @@
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
M=icassoGet(sR,'M');
rdim=icassoGet(sR,'rdim');
clusterparameters={'simfcn','abscorr','s2d','sim2dis','strategy','AL','L','rdim'};
Wt = cell2mat(sR.W');
similarity = abs(corr(Wt'));
clusterparameters={'simfcn',similarity,'s2d','sim2dis','strategy','AL','L','rdim'};
num_of_args=length(clusterparameters);

%%%%%%%%%%%%% arguments %%%%%%%%%%%%%%%%
Expand Down
24 changes: 17 additions & 7 deletions relica_plots.m
View file
Original file line number Diff line number Diff line change
Expand Up @@ -26,19 +26,29 @@
% EEG - Output dataset: RELICA data is in EEG.etc.RELICA, the same is
% saved in folder_relica folder
%
% Author: Fiorenzo Artoni, The Biorobotics Institute / EPFL, 2017 %
% Author: Dr. Fiorenzo Artoni, 2019 %
%
% Reference:
% References:
% (1) Artoni, F., Menicucci, D., Delorme, A., Makeig, S., & Micera, S. (2014).
% RELICA: a method for estimating the reliability of independent components.
% NeuroImage, 103, 391-400.


% Copyright (C) 2017 Fiorenzo Artoni, The Biorobotics Institute , EPFL, SCCN, [email protected]
% NeuroImage, 103, 391-400.
%
% (2) Artoni, F., Delorme A., Makeig S. (2018)
% Applying dimension reduction to EEG data by Principal Component Analysis
% reduces the quality of its subsequent Independent Component
% decomposition, Neuroimage 175 176-187
%
% This project was in part supported by the European Union's Horizon 2020
% research and innovation programme under Marie Sklodowska-Curie Action
% agreement no. 750947 (BIREHAB)
%
% Acknowledgments go to Ramon Martinez-Cancino (SCCN/INC/UCSD 2019) for making the
% algorithm available and parallelized on the NSG server.
% Acknowledgments go also to Arnaud Delorme and Scott Makeig (SCCN/INC/UCSD 2019)
% for the precious inputs and ideas to perfect the project.
% Clustering and relative visualization within RELICA makes use of modified
% routines from J. Himberg's open source FastICA - ICASSO package
% Beamica is part of C. Kothe's open source BCILAB toolbox
% Beamica is part of C. Kothe's open source BCILAB toolbox %
%
% This program is free software; you can redistribute it and/or modify
% it under the terms of the GNU General Public License as published by
Expand Down

0 comments on commit bdc39e8

Please sign in to comment.