Functional Connectivity (Cross-Correlation Method) - Revision history

Chris: /* Censor Data (Optional Step) */

2022-04-27T21:37:08Z

Censor Data (Optional Step)

← Older revision		Revision as of 17:37, 27 April 2022
Line 41:		Line 41:
	for i=1:length(Zprime)		for i=1:length(Zprime)
	temp=Zprime{i}		temp=Zprime{i}
	temp>threshold=replacement;		temp(temp>threshold)=replacement;
	temp<-threshold=-1*replacement;		temp(temp<-threshold)=-1*replacement;
	Zprime{i}=temp;		Zprime{i}=temp;
	end		end

Chris: /* Censor Data (Optional Step) */

2022-04-27T21:36:15Z

Censor Data (Optional Step)

← Older revision		Revision as of 17:36, 27 April 2022
Line 38:		Line 38:
	replacement=NaN; %could instead have used replacement==thresh		replacement=NaN; %could instead have used replacement==thresh
	thresh=2.5; %could use a more-or-less stringent threshold		thresh=2.5; %could use a more-or-less stringent threshold
	for i=1:length(Z)		Zprime=Z
	temp=Z{i}		for i=1:length(Zprime)
			temp=Zprime{i}
	temp>threshold=replacement;		temp>threshold=replacement;
	temp<-threshold=-1*replacement;		temp<-threshold=-1*replacement;
	Z{i}=temp;		Zprime{i}=temp;
	end		end
	If you had an outside indicator of data quality, you could similarly use that to identify data to be removed. For example, a list of volumes with excessive motion (as determined during functional data realignment, or the ArtRepair toolbox) could be read in:		If you had an outside indicator of data quality, you could similarly use that to identify data to be removed. For example, a list of volumes with excessive motion (as determined during functional data realignment, or the ArtRepair toolbox) could be read in:

Chris: /* Censor Data (Optional Step) */

2022-04-27T21:35:25Z

Censor Data (Optional Step)

← Older revision		Revision as of 17:35, 27 April 2022
Line 39:		Line 39:
	thresh=2.5; %could use a more-or-less stringent threshold		thresh=2.5; %could use a more-or-less stringent threshold
	for i=1:length(Z)		for i=1:length(Z)
	Z{i}>threshold=replacement;		temp=Z{i}
	~~Z{i}~~<-threshold=-1*replacement;		temp>threshold=replacement;
			temp<-threshold=-1*replacement;
			Z{i}=temp;
	end		end
	If you had an outside indicator of data quality, you could similarly use that to identify data to be removed. For example, a list of volumes with excessive motion (as determined during functional data realignment, or the ArtRepair toolbox) could be read in:		If you had an outside indicator of data quality, you could similarly use that to identify data to be removed. For example, a list of volumes with excessive motion (as determined during functional data realignment, or the ArtRepair toolbox) could be read in:

Chris at 00:23, 18 June 2019

2019-06-18T00:23:14Z

← Older revision		Revision as of 20:23, 17 June 2019
Line 1:		Line 1:
	If you consider two vectors containing ''t'' values representing neural activity over ''t'' time points, the degree to which the values in these vectors move together, as indexed by the correlation between these vectors, can be taken as a measure of functional connectivity. The underlying assumption is that the activity in two regions should be related if there is some sort of underlying connection between them.		If you consider two vectors containing ''t'' values representing neural activity over ''t'' time points, the degree to which the values in these vectors move together, as indexed by the correlation between these vectors, can be taken as a measure of functional connectivity. The underlying assumption is that the activity in two regions should be related if there is some sort of underlying connection between them.

	This page assumes you have already obtained time series matrices using the approaches [[Time_Series_Analysis \| described here]]. If these data were obtained in FreeSurfer surface space, there will be 2 matrices (lh, rh) times 'r' runs. ~~As I~~ have ~~not yet worked out the process for voxel-~~space ~~(i.e.~~, ~~using SPM), details are sketchy, but it seems probable that there would only be~~ a ~~single~~ matrix ~~of values per run~~. In either case, some data cleaning may be required.		This page assumes you have already obtained time series matrices using the approaches [[Time_Series_Analysis \| described here]]. If these data were obtained in FreeSurfer surface space, there will be 2 matrices (lh, rh) times 'r' runs. If subcortical activations have been obtained in MNI305 space, a third matrix will also be available. In either case, some data cleaning may be required.

	These steps will be carried out in MATLAB. The first step then would be to start up MATLAB in your terminal (the ampersand lets the program run in the background so that you can continue to use your terminal window to do other things if you wish):		These steps will be carried out in MATLAB. The first step then would be to start up MATLAB in your terminal (the ampersand lets the program run in the background so that you can continue to use your terminal window to do other things if you wish):

Chris: /* Data Cleaning */

2018-08-14T16:43:21Z

Data Cleaning

@@ Line 13: / Line 13: @@
 == Data Cleaning ==
 Steps undertaken during earlier fMRI data conversion and/or preprocessing may influence the sorts of steps you wish to perform in order to clean up your data:
 === Normalize Data ===
@@ Line 61: / Line 38: @@
   replacement=NaN; %could instead have used replacement==thresh
   thresh=2.5; %could use a more-or-less stringent threshold
-  zrun05(zrun05>thresh)=replacement;
+  for i=1:length(Z)
- zrun05(zrun05<-thresh)=-replacement;
+   Z{i}>threshold=replacement;
+   Z{i}<-threshold=-1*replacement;
 If you had an outside indicator of data quality, you could similarly use that to identify data to be removed. For example, a list of volumes with excessive motion (as determined during functional data realignment, or the ArtRepair toolbox) could be read in:
   bad_vols=load('art_censored.txt')
@@ Line 72: / Line 50: @@
-Note that the above hypothetical example would assume that all the volumes were present in the data. If the initial ''n'' volumes have already been dropped from the data, then you would need to remove from consideration the first ''n'' volumes and then subtract ''n'' from the remaining bad_vols. This is because after dropping the first 6 volumes, what was originally the 7th time point is now actually the first time point in the remaining time series:
+Note that the above hypothetical example would assume that all the volumes were present in the data. If the initial ''n'' volumes have already been dropped from the data, then you would need to remove from consideration the first ''n'' volumes and then subtract ''n'' from the remaining bad_vols. This is because after dropping the first 4 volumes, what was originally the 5th time point is now actually the first time point in the remaining time series.
 == Know What's What ==

Chris: /* Normalize Data */

2018-05-18T03:08:57Z

Normalize Data

← Older revision		Revision as of 23:08, 17 May 2018
Line 43:		Line 43:
	The normalized matrix will have z-scores for each time point, calculated using the mean and standard deviation for each column (region) separately. Thus, a normalized value of 1 in Region A will mean that the activation value was 1 standard deviation above the average for that region.		The normalized matrix will have z-scores for each time point, calculated using the mean and standard deviation for each column (region) separately. Thus, a normalized value of 1 in Region A will mean that the activation value was 1 standard deviation above the average for that region.

	You may be surprised to learn that a MATLAB function, <code>normalizeMatrix.m</code> has been created to facilitate the normalization of our time series data. If it has been copied to your path, you can easily normalize your multidimensional matrix or cell array '''''M''''' (recall that '''''M''''' is a cell array if you used the <code>loadFSTS.m</code> function). The built-in MATLAB function, <code>zscore()</code> unfortunately doesn't work on cell arrays. The <code>normalizeMatrix.m</code> function was written to permit seamless normalization of the time series matrices, whether they are 3D matrices or cell arrays. Additional functionality was added to drop rows (i.e., time points) specified in an array, thus eliminating the need to drop specific time points in a separate step as described above.		You may be surprised to learn that a MATLAB function, <code>normalizeMatrix.m</code> has been created to facilitate the normalization of our time series data. If it has been copied to your path, you can easily normalize your multidimensional matrix or cell array '''''M''''' (recall that '''''M''''' is a cell array if you used the <code>loadFSTS.m</code> function). The built-in MATLAB function, <code>zscore()</code> unfortunately doesn't work on cell arrays. The <code>normalizeMatrix.m</code> function was written to permit seamless normalization of the time series matrices, whether they are 3D matrices or cell arrays.

			Additional functionality was added to drop rows (i.e., time points) specified in an array, thus eliminating the need to drop specific time points in a separate step as described above. '''You should only do this if volumes were not already dropped from the .nii files prior to importing the timeseries!'''. If you drop 4 rows from the .nii files, and then drop the first four rows again, the first 8 volumes will have been dropped from your time series! However even if you have already dropped volumes from the start of the .nii file, you may wish to drop volumes from the end of the time series (e.g., if the experiment ended before the run was finished).
	droprows=[-2 -1 0 1 2 3]; %omit rows 1, 2, 3 and end, end-1 and end-2		droprows=[-2 -1 0 1 2 3]; %omit rows 1, 2, 3 and end, end-1 and end-2
	Z=normalizeMatrix(M, droprows);		Z=normalizeMatrix(M, droprows);

Chris: /* Load Your Time Series Data */

2017-11-22T01:59:49Z

Load Your Time Series Data

@@ Line 10: / Line 10: @@
 == Load Your Time Series Data ==
 The general instructions for importing your detrended time series data into MATLAB can be found [[Importing_Time_Series_Into_MATLAB | here]]
 == Data Cleaning ==

Chris: /* Load Your Time Series Data */

2017-11-22T01:59:16Z

Load Your Time Series Data

← Older revision		Revision as of 21:59, 21 November 2017
Line 9:		Line 9:

	== Load Your Time Series Data ==		== Load Your Time Series Data ==
	The general instructions for importing your detrended time series data into MATLAB can be found [[Importing_Time_Series_Into_MATLAB here]]		The general instructions for importing your detrended time series data into MATLAB can be found [[Importing_Time_Series_Into_MATLAB \| here]]

	The next step is to load your data matrices. Assuming the time series are plaintext files containing only structured numerical data (e.g., as produced by the gettimecourses.sh script), you can use the <code>load</code> function to import these data.		The next step is to load your data matrices. Assuming the time series are plaintext files containing only structured numerical data (e.g., as produced by the gettimecourses.sh script), you can use the <code>load</code> function to import these data.

Chris at 01:58, 22 November 2017

2017-11-22T01:58:58Z

← Older revision		Revision as of 21:58, 21 November 2017
Line 9:		Line 9:

	== Load Your Time Series Data ==		== Load Your Time Series Data ==
			The general instructions for importing your detrended time series data into MATLAB can be found [[Importing_Time_Series_Into_MATLAB here]]

	The next step is to load your data matrices. Assuming the time series are plaintext files containing only structured numerical data (e.g., as produced by the gettimecourses.sh script), you can use the <code>load</code> function to import these data.		The next step is to load your data matrices. Assuming the time series are plaintext files containing only structured numerical data (e.g., as produced by the gettimecourses.sh script), you can use the <code>load</code> function to import these data.
	=== Functionally-assisted loading ===		=== Functionally-assisted loading ===

128.205.172.22: /* Partition Time Series */

2017-11-14T20:15:46Z

Partition Time Series

← Older revision		Revision as of 16:15, 14 November 2017
Line 194:		Line 194:
	bookends(5,1):bookends(5,2)], :);		bookends(5,1):bookends(5,2)], :);
	%You can similarly use the returned '''conditions''' matrix to index the block boundaries for specific block conditions		%You can similarly use the returned '''conditions''' matrix to index the block boundaries for specific block conditions

			<em>Note:</em> as of today (November 14, 2017), the findBlockBoundaries function doesn't adjust for dropped volumes. We now typically drop the first 4 volumes from our time series before any analyses are done. This means that the time points recorded in the source .mat files will be off by the number of dropped volumes. Until the findBlockBoundaries function is modified to take the number of volumes dropped as a parameter, the fix will be to subtract d from the bookends arrays using <code>cellfun()</code>:
			ndropped=4;
			new_bookends=cellfun(@(x) x-ndropped, bookends, 'UniformOutput', false);

	== Save Your Data ==		== Save Your Data ==