[Homer-users] Using Matlab to substract conditions for ROI

[Theodore Huppert]

Felix-

    I tried to repeat the steps you had done, but I was unable to  
reproduce the error. I would be happy to take a look at the data file  
if you wish to send me (thuppert at nmr.mgh.harvard.edu) the *.hmr saved  
data file.  I can see if I can track down and fix any errors.

Also,note that the analysis in HOMER is done at three levels: i)  
individual scans ii) scan sessions and iii) multiple scans/subjects.   
The "AVERAGE" button computes the average across only a single scan  
(level i).  The "AVERAGE ALL" button computes the average across a  
single session (level ii).  The ROI average button computes TWO the  
region-of-interest averages for the roi in level II and the group  
average (level III).  These should be stored in the *.hmr file as DOT 
{idx}.ROI for level 2 and DOT{idx}.ROIavg for level 3.    If you want  
an ROI average PER condition, you should put each condition into a  
seperate "Session" (level 2 analysis).  Otherwise, the average will  
be computed across all the files.  A seperate MeasurementList can be  
specified for each condition (to my knowledge this works... it worked  
for me just now let me know if you find otherwise).

Note, there is also a way to specifiy more then 1 condition within a  
single file.  In the *.nirs files, the stimulus timing is specified  
by the variable "s".  This is typically a vector of 0's and 1's  
marking the onsets of the stimuli  (or it can be a recorded signal  
and HOMER will do some rising edge detection to find the onsets).   
However, this variable (s) can also be a matrix, where each column  
specifies a different stimulus condition.  I.e. s(:,1) for condition  
1, s(:,2) for condtion 2, etc.  Once loaded into HOMER, these  
additional conditions can be added to the linear model using the  
"Averaging Options" button and adding these conditions as  
regressors.  This is also how physiology measures can be added as  
regressors to the model.  See the sample file named  
"PhysiologyPhantom.nirs" for an example.

-Ted

Theodore Huppert, Ph.D.
thuppert at nmr.mgh.harvard.edu




On Aug 21, 2007, at 1:31 PM, 李振国 wrote:

> hello,
> Thanks for your letter, I didn't have the case like this ,sorry!
> thank you
> guozi li
>
>
>
> 在2007-08-21，"Felix Le" <spiderbo100 at hotmail.com> 写道：
>
> Hi,
>
> I've done all the steps required to analyse an ROI and saved  
> different files for each condition I want to analyze. But when I  
> open the files in Matlab, I only get data average (AvgC) in the  
> DataAvg section. When I go in the ROI section, the AvgC matrix is  
> full of 0 and -0, wich is weird.
>
> Can I use the DataAvg for my substractions based on the specific  
> MeasListAct I've set up ?
>
> Thank you,
>
> Felix
>
> Discover the new Windows Vista Learn more!
>
>
> 全 国 万 名 职 场 女 性，齐 聚 非 常 美 女，惊 艳  
> 2 0 0 7
> _______________________________________________
> Homer-users mailing list
> Homer-users at nmr.mgh.harvard.edu
> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/homer-users




-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.nmr.mgh.harvard.edu/pipermail/homer-users/attachments/20070822/b375d13c/attachment.html