[Homer-users] Short Separation in Developer's Version

[Meryem Ayse Yucel]

Hi Rachel,

1) Your s vector is empty. You need to fix it.

2) Your source and detector positions does not seem to be either cm or mm.
The smallest distance calculated by homer is around 110 for your SD
geometry, so if you type 120 for ssthreshold, it will perform short
separation regression since it will find two SD pairs which have shorter
distance. You need to fix your units.

Hope this helps.

Meryem




> Based on this error code, it would appear that I need to use the 1.5 cm;
> however, the SS channels do not flatten and I get the message: There are
> no
> short separation channels in this probe...performing regular
> deconvolution.
> Design matrix is VERY poorly scaled...(RCond=0.000000e+00), cannot perform
> computation
> There are no short separation channels in this probe...performing regular
> deconvolution.
> Design matrix is VERY poorly scaled...(RCond=0.000000e+00), cannot perform
> computation
>
> I am attaching a picture of my probe array after running the processing
> stream with SS. SS channels were 1 cm, all other channels were 3 cm.
>
>
>
> On Wed, Apr 29, 2015 at 3:13 PM, Rachel Mulheren <mulherrw at dukes.jmu.edu>
> wrote:
>
>> When the processing gets "stuck" on the SS, this is the following error
>> code I receive:
>>
>> Reference to non-existent field 'svs'.
>>
>> Error in EasyNIRS_ProcessOpt>pushbutton_Callback (line 268)
>> eval( sprintf( '%s_result( %s );',
>> procInput.procFunc.funcName{eventdata},
>> sargin(2:end) ) );
>>
>> Error in gui_mainfcn (line 96)
>>         feval(varargin{:});
>>
>> Error in EasyNIRS_ProcessOpt (line 42)
>>     gui_mainfcn(gui_State, varargin{:});
>>
>> Error in
>>
>> EasyNIRS_ProcessOpt>EasyNIRS_ProcessOpt_OpeningFcn/@(hObject,eventdata)EasyNIRS_ProcessOpt('pushbutton_Callback',hObject,3,guidata(hObject))
>>
>>
>> Error while evaluating uicontrol Callback
>>
>> Reference to non-existent field 'svs'.
>>
>> Error in EasyNIRS_ProcessOpt>pushbutton_Callback (line 268)
>> eval( sprintf( '%s_result( %s );',
>> procInput.procFunc.funcName{eventdata},
>> sargin(2:end) ) );
>>
>> Error in gui_mainfcn (line 96)
>>         feval(varargin{:});
>>
>> Error in EasyNIRS_ProcessOpt (line 42)
>>     gui_mainfcn(gui_State, varargin{:});
>>
>> Error in
>>
>> EasyNIRS_ProcessOpt>EasyNIRS_ProcessOpt_OpeningFcn/@(hObject,eventdata)EasyNIRS_ProcessOpt('pushbutton_Callback',hObject,3,guidata(hObject))
>>
>>
>> Error while evaluating uicontrol Callback
>>
>> Index exceeds matrix dimensions.
>>
>> Error in hmrDeconvTB_SS3rd (line 141)
>>         Ass = y(:,conc,mlSSlst(iSS));
>>
>> Error in EasyNIRS_Process (line 88)
>>         eval( fcall );
>>
>> Error in EasyNIRS>runNIRSsignalProcess (line 1022)
>> [procResult procInput err] = EasyNIRS_Process(hmr);
>>
>> Error in EasyNIRS>popupmenuNIRSsignalProcess_Callback (line 1435)
>>         runNIRSsignalProcess(iGrp,iSubj,iRun,handles);
>>
>> Error in gui_mainfcn (line 96)
>>         feval(varargin{:});
>>
>> Error in EasyNIRS (line 43)
>>     gui_mainfcn(gui_State, varargin{:});
>>
>> Error in
>>
>> @(hObject,eventdata)EasyNIRS('popupmenuNIRSsignalProcess_Callback',hObject,eventdata,guidata(hObject))
>>
>>
>> Error while evaluating uicontrol Callback
>>
>> >>
>>
>> On Wed, Apr 29, 2015 at 12:09 PM, David Boas <dboas at nmr.mgh.harvard.edu>
>> wrote:
>>
>>>  Most likely the units are mm and you need to use 15 for the
>>> rhoSD_SSThresh
>>>
>>>  The fact that the processing gets “stuck” when you use 15 probably
>>> indicates that it is busy calculating the short separation regression.
>>> Let
>>> me verify, you say it gets “stuck”, but it doesn’t crash?
>>>
>>>  What type of computer are you using? How big is your data set?
>>>
>>>
>>>
>>>  On Apr 29, 2015, at 11:49 AM, Rachel Mulheren <mulherrw at dukes.jmu.edu>
>>> wrote:
>>>
>>>    Hello,
>>>
>>>  I included 2 short separation channels (1 cm) in my probe and now that
>>> I
>>> am running the processing stream including the parameters for short
>>> separation, I get the following message: There are no short separation
>>> channels in this probe...performing regular deconvolution.
>>> Design matrix is VERY poorly scaled...(RCond=0.000000e+00), cannot
>>> perform computation
>>>
>>>  Eyeballing the HRF after running this processing stream, I am not
>>> seeing
>>> a difference with and without the short separation, so I am guessing
>>> that
>>> it was not extracted from the signal.
>>>
>>>  Suggestions for what I might be doing incorrectly?
>>>
>>>  I am using Block Avg Short Sep
>>>  trange -5 to 17
>>>  gstd 1
>>>  gms 1
>>>  rhoSD_SSThresh 1.5
>>>
>>>  I have also tried with rhoSD_SSThresh 15 (in case there was a mix-up
>>> with mm and cm), but the processing starts and gets "stuck" on the
>>> short
>>> separation.
>>>
>>>  Thanks,
>>>
>>>  Rachel
>>> _______________________________________________
>>> Homer-users mailing list
>>> Homer-users at nmr.mgh.harvard.edu
>>> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/homer-users
>>>
>>>
>>>
>>
> _______________________________________________
> Homer-users mailing list
> Homer-users at nmr.mgh.harvard.edu
> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/homer-users