latest update problem

[astrid]

Hi,
I updated Lead on Friday (v 1.6.3.4) and now i cannot open 3D reconstructions I did with the previous version anymore, either in individuals or in the group, here are the error msg:
Error using ea_error (line 15)
Couldn’t visualize electrode from patient 1.

Error in ea_elvis (line 104)
ea_error([‘Couldn”t visualize electrode from patient ‘,num2str(pt),’.’]);

Error in ea_write (line 46)
resultfig=ea_elvis(options);

Error in ea_autocoord (line 200)
ea_write(options)

Error in ea_run (line 95)
ea_autocoord(options);

Error in lead_dbs>run_button_Callback (line 190)
ea_run(‘run’,options);

Error in gui_mainfcn (line 95)
feval(varargin{:});

Error in lead_dbs (line 43)
gui_mainfcn(gui_State, varargin{:});

Error in @(hObject,eventdata)lead_dbs(‘run_button_Callback’,hObject,eventdata,guidata(hObject))

Error while evaluating UIControl Callback

>>
Error using ea_error (line 15)
Couldn’t visualize electrode from patient 1.

Error in ea_elvis (line 104)
ea_error([‘Couldn”t visualize electrode from patient ‘,num2str(pt),’.’]);

Error in lead_group>vizbutton_Callback (line 362)
resultfig=ea_elvis(options,M.elstruct(get(handles.patientlist,’Value’)));

Error in gui_mainfcn (line 95)
feval(varargin{:});

Error in lead_group (line 42)
gui_mainfcn(gui_State, varargin{:});

Error in @(hObject,eventdata)lead_group(‘vizbutton_Callback’,hObject,eventdata,guidata(hObject))

Error while evaluating UIControl Callback

I guess there is an issue with the latest update?
Best,
Astrid

[ningfei]

Hi Astrid,

The above log seems to be from lead group. Have you also tried it in lead dbs with the data from patient 1 (the first one in lead group subject list)? Before doing that, please apply a hotfix first. Just unzip it into your lead folder and overwrite the old files. You may also want to reset the lead preferences to default from the Preferences menu. Let me know what you got then.

Best,
Ningfei

[astrid]

With these options, there is still the same bugs with electrode visualization,
when investigating more, the error seems to come from ea mapelmode2reco:

[el_render(pt).el_render,el_label(:,pt)]=ea_showelectrode(resultfig,elstruct(pt),pt,options);
To use ‘radtodeg’, you might need:
radtodeg – Mapping Toolbox

Error in ea_mapelmodel2reco (line 61)
if 90-radtodeg(acos(dot(…

Error in ea_showelectrode (line 56)
[X,electrode,err]=ea_mapelmodel2reco(options,elspec,elstruct,side,resultfig);

Best,
Astrid

[andreashorn]

Hi astrid,

could you try to replace “radtodeg” with “rad2deg” in line 61 in ea_mapelmodel2reco?

Best, Andy

[astrid]

Hi Andreas,
I did but also got the pbm as rad2deg is also a function of the Mapping Toolbox
Best!
Astrid

[andreashorn]

Hi Astrid,

true, please replace with this file:
https://github.com/leaddbs/leaddbs/blob/master/ea_mapelmodel2reco.m

Best, Andy

[astrid]

Hi,
thank you for the answers! it is working fine now,
I have an additional issue though concerning the connectomes, the groupconnectome file seems to have changed path, I unzipped the lead data file as explained, but it still did not find any modality (indicating that no connectivity data is found)
I tried a dirty way of doing this by renaming groupconnectomes data.mat and placing them in a folder in dMRI and it seemed to work (at least it built the ea_pm) untill a certain point,
finally i wondered how in addition to the ea_pm we could get the meshes of the fibers to plot them in trackvis for instance?
Thank you again!
Best,
Astrid

[ningfei]

Hi Astrid,

Yes. Both the path and the naming of the group connectome data have changed. It should be placed in lead/connectomes/dMRI/ as you already noticed.

Maybe I don’t get you right, but if you just want to visualize the fibers from groupconnectome data in trackvis, I guess Andy might already have the trk files. In general, the conversion can be done by ea_ftr2trk, if the space is known.

Best,
Ningfei

[pdeman]

hi,
in the connectome gui, stimulation-based results part.
the list of modality available correspond to what ? the group connectomes ?

to fill this list, you use this:

% check for canonical fiber sets
fdfibs=dir([ea_getconnectomebase('dmri'),filesep]);
for fdf=1:length(fdfibs)
    if fdfibs(fdf).isdir && ~strcmp(fdfibs(fdf).name(1),'.')
        [~,fn]=fileparts(fdfibs(fdf).name);
        modlist{cnt}=fn;
        sf(cnt)=1;
        cnt=cnt+1;
    end
end

ea_getconnectomebase return ‘lead/connectomes/dMRI/’ (and then you add again a filesep it’s not necessary)
then you list what’s inside this directory using dir and you keep in modlist only the directory (if fdfibs(fdf).isdir)
so the connectome files are not seen.

we made a directory (let’s say “general”) in ‘/lead/connectomes/dMRI’ and then we see this directory in the modality list in the stimulation based results.

then when we try to do something.
it looks for a data.mat file in /lead/connectomes/dMRI.
what should be this data.mat file ?

moreover we would like to generate for each patients, the fiber meshs resulting from the stimulation-based results.
and load in lead_group each patient fiber mesh (not the total fibers but the one selected by the stimulation-based results).
if each patient can have their fibers in a different color that’s a bonus:)

[ningfei]

Hi,

Yes, the filesep there is reduntant, but this won’t affect the result of dir.

As I said, the groupconnectome data should be unzipped into lead/connectomes/dMRI/ now. You will then see some folders with the name like “Groupconnectome (Horn 2013) full” there. The data.mat files are in these folders. If you have all the group connectome data ready, you will see them in the “Prepare DBS stats” section under “which connectivity metric to use” (seems the popupmenu is not large enough to show the full names of the folders), from where you can also choose the patient specific fiber tracts.

Best,
Ningfei

[andreashorn]

… so new structure should be
/dMRI/NameOfGroupConnectome/data.mat

…as a side note, I had written this change to you guys when I implemented it since I knew it would lead to trouble :)
Not sure whether via mail or slack but I did think of you when changing the structure.

The change was/will be needed because we now have a definite format for structural and functional connectome files. In fMRI, connectomes are not stored as single files but folders with a multitude of subject files. I also wanted to allow different ways to store structural connectomes that need multiple files. So I decided to replace the .mat files with folders inside of which is (for now) a single file named data.mat

Best, Andy

[pdeman]

ok for the folder problem.

about the display of the fibertracks selected by the vat for each patient.
is it possible to display all of them in lead_group ?
we can’t find the mesh of the fibertracks selected.
we have the ea_pm file, but not the fibers.
(I am talking specifically about the fibers selected by the simulation of the stimulations, not the DTI of the patient, we don’t have them for these patients, we use the groupconnectom).

• This reply was modified 7 years, 7 months ago by pdeman.

[andreashorn]

As said via email, no (see mail).
Please use a fibertracking software for that (TrackVis/DSI Studio)

[Author]

Posts