[Homer3/AtlasViewerGUI] Homer3 compatibility for AtlasViewer

[Meenakshi Siddharthan]

Dear all,

I started this thread to ask if Homer3 output can be used to produce ImageRecon in AtlasViewer. With the latest version of Homer3 (updated 25th July) I get this error as shown in the screenshot.

I am following updates on GitHub and will try again with AtlasViewer when its updated. Do let me know if there is anything else I should try to get ImageRecon in AVGUI to work meanwhile!

Thank you!

[Boas, David]

We hope to have this updated this week. If you don’t hear back this week, then please post again to this thread.

--
You received this message because you are subscribed to the Google Groups "Homer-fNIRS and AtlasViewer Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email to homer-fnirs...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/homer-fnirs/5a77db91-ade1-46df-87fd-a7a75af2b6c8o%40googlegroups.com.

[may...@bu.edu]

Dear Alicia MG and Meenan S,

Just checked the data Alicia sent, it still does not have the expected wavelength order, which is the cause of the error below I believe:

The order should all measurements from first wavelength in the first half of ml (and so d too!) and so on. Since you are working with homer2, you need to fix this. Homer3 will be modified to remove this rigid requirement. Hope this helps!

Index in position 1 exceeds array bounds (must not exceed 30).

Error in ImageRecon>image_recon_Callback (line 338)

yavgimg = yavgimg(lstLS_all,:);

Error in gui_mainfcn (line 95)

        feval(varargin{:});

Error in ImageRecon (line 16)

    gui_mainfcn(gui_State, varargin{:});

Error in

matlab.graphics.internal.figfile.FigFile/read>@(hObject,eventdata)ImageRecon('image_recon_Callback',hObject,eventdata,guidata(hObject)) 

[Alicia]

Dear Meryem,

Appreciate you looking into this. We had previously tried this and had done it to get GLM but had not changed the 'd'.

We have now re-ordered "SD.MeasList" & "d" and the recon has worked for some of our data but not all , which is strange (all the .nirs files should be all the same after reordering). 

We are having a look to try and figure this out and also going to try it in Homer/Atlas 3 now that it is updated,  and see if it resolves the issue. 

Will let you know.

Thank you again :).

Best wishes,

Alicia

[Meenakshi Siddharthan]

Hi all,

I've just tried with the latest version of Homer3 and AtlasViewerGUI to see if I can get ImageRecon. However, I seem to be getting the same error as prior (attached screenshot).

I believe it's due to groupresults.mat that is produced by Homer3 being different to the one was produced by Homer2. It seems the output (procResult) is now in a seperate mat file to groupResults which should be why I am unable to get the ImageRecon working. 

Do you recommend anything we could do on our side to get it working? Thank you so much for your help so far!

Regards,

Meena

[Boas, David]

Jay is working on this issue right now. I hope he will have it fixed this week.

Please check back at the end of the week if you don’t hear from us.

 

 

From: <homer...@googlegroups.com> on behalf of Meenakshi Siddharthan <meena.sidd...@gmail.com>

Reply-To: "homer...@googlegroups.com" <homer...@googlegroups.com>
Date: Monday, August 3, 2020 at 5:16 AM
To: Homer-fNIRS and AtlasViewer Community <homer...@googlegroups.com>

--

You received this message because you are subscribed to the Google Groups "Homer-fNIRS and AtlasViewer Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email to homer-fnirs...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/homer-fnirs/b23a6163-6b13-461d-b6e7-afa33663816co%40googlegroups.com.

[Stephen Tucker]

Hi all,

The latest version of AtlasViewer on GitHub should be compatible with the Homer3 output. It has been tested but may have a few issues, please report them here! Thanks.

Stephen

[Alicia]

Hi Meryem,

We have still not been able to get the recon to work with all our files after re ordering both SD MeasList and d. However, using a different folder format we have been able to get it to work but only with 1 or 2 files and not others.

I have attached the steps we used to re order d. We made the assumption that d is structured as 1 2 1 2 being wavelength 1 first and wavelength 2 second, but unlike SD MeasList it does not actually specify. I have attached the re ordering steps we used. 

We will also do a run through on Atlas Viewer 3 now it has been updated. However, as all our data has already been analysed using Homer 2, it would be good to work out the Atlas Viewer reordering issue on 2 as well. 

Thank you!

Best wishes,

Alicia 

[Meenakshi Siddharthan]

Hi all,

Thank you for updating both Homer and AVgui! I've just tried out the latest version of Homer3 and of AtlasViewer (downloaded 31st Aug). I did as follows on MatLabR2020a:

1. Reorganized data as suggested before so SD.Measlist from 1 2 1 2 to 1 1 2 2 and d (assuming it is also arranged as 1 2 1 2) from 1 2 1 2 to 1 1 2 2 

2. Direct to Homer3-master folder > setpaths > Open Homer3 > Direct to folder with .nirs file > run processing stream on .snirf file

3. Direct to AtlasViewer-master folder > setpaths > Open AtlasViewerGUI > Direct to the folder containing the .nirs, .snirf, groupResults.mat, Subj_test.mat, test.mat and .digpts files

I'm able to complete steps 1-2 without a hitch. However when I attempt to open the results of Homer3 in AVgui I get an error message:

Unrecognized function or variable 'ConfigFileClass'.

Error in RegistriesClass (line 43)

            cfg = ConfigFileClass();

Error in ProcStreamClass (line 31)

            obj.reg = RegistriesClass();

... and so on

I'm unable to attach a screen shot of the error message so these are just the first few lines.With the error message, I'm not able to get a head model loaded onto the GUI (so it's just an empty space). 

I tried running the same .nirs file through Homer2 and loaded it onto the latest AVgui but got the same error. Additionally, I tried running the .nirs file through Homer2 and then loading it onto Homer2's AVgui and was able to get the head model with the digpts on as usual. 

So the .nirs file should not be the issue I believe. Do let me know if I should attempt any other steps or if you require any further detail about this! Thank you!

[Yucel, Meryem]

Dear Alicia,

I recall you had wavelengths interleaved in your nirs files i.e. in measlist you had the last column (1 2 1 2 and so on). So the code you shared won't fix it for d, you can try this:

SD.MeasList = sortrows(SD.MeasList, 4)

d0 = d;

d0(:,1:size(d0,2)/2) = d(:,1:2:end);

d0(:,size(d0,2)/2+1:end) = d(:,2:2:end);

d = d0;

Hope this helps.

Meryem

---

From: homer...@googlegroups.com <homer...@googlegroups.com> on behalf of Alicia <aliciago...@gmail.com>
Sent: Sunday, August 23, 2020 10:18 AM

To: Homer-fNIRS and AtlasViewer Community <homer...@googlegroups.com>
Subject: Re: [Homer3/AtlasViewerGUI] Homer3 compatibility for AtlasViewer

--

You received this message because you are subscribed to the Google Groups "Homer-fNIRS and AtlasViewer Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email to homer-fnirs...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/homer-fnirs/46ed59eb-8e8f-479c-88f6-eaeed1621651o%40googlegroups.com.

[Meenakshi Siddharthan]

Dear Meryem,

Thank you for the reply! So we've rearranged our .nirs files using the matlab code you sent us and are now able to get our recon figures using Homer2!

May I ask if we require a Probe.SD file to get image recon for group results? Is it possible to just use the mean of the participants' digpts files to obtain group image recon? 

Also can I just confirm that the units on the color scale of the image recon diagram are µM?

Thank you!