search

petmri / ROCKETSHIP

Dynamic Contrast Enhanced (DCE) MRI Analysis Software
GNU General Public License v2.0
48 stars 14 forks source link

dce_run_A #4

Closed doctorbulat closed 8 years ago

doctorbulat commented 8 years ago

Hello again! I tried to run a. But have some troubles again. "Input image sizes not the same".

And second question about AIF/RR. How to make a binary mask, with 1 in the AIF/RR region and zero elsewhere; or a T1 map, where the AIF/RR region contains valid T1 values (in ms), and all other regions are zero? which programm should I use? Is there any manuals how to do them! And also how to select ROI? I think the solution is the same? Thank you very much and sorry for so many problems! Best regards, Bulat.

doctorbulat commented 8 years ago

I tried even to input coregistrated in spm images, but still i have "Input image sizes not the same"...

ironictoo commented 8 years ago

The input images have to be exactly the same size, there is no getting around that, they should be acquired that way in the first place. If they are off a little for some reason you can resample them with a program like ImageJ, that would probably be the easiest thing to do. But you need to make sure the anatomy lines up precisely if it is a little off you will get artifacts were anatomic structures don't line up. To make the binary masks we used ImageJ, I recommend getting FIJI which is ImageJ with a bunch of nice plugins and an updater. ImageJ can also be used to make ROI files.

On Saturday, January 30, 2016 8:07 AM, doctorbulat <notifications@github.com> wrote:

Hello again! I tried to run a But have some troubles again "Input image sizes not the same" I checked the matrix size for dynamic datasets It is equal to 192x132 And matrix size t1_pre_flip_2deg and t1_pre_flip_15deg (images to get t1_map) It is equal to 192x138 Not a big difference between 132 and 138 Is that the solution for the problem? the matrix sizes must be absolutely the same?And second question about AIF/RR How to make a binary mask, with 1 in the AIF/RR region and zero elsewhere; or a T1 map, where the AIF/RR region contains valid T1 values (in ms), and all other regions are zero? which programm should I use? Is there any manuals how to do them! And also how to select ROI? I think the solution is the same? Thank you very much and sorry for so many problems! Best regards, Bulat— Reply to this email directly or view it on GitHub.

ironictoo commented 8 years ago

You going to have to be more specific about what problem you are having.

On Monday, February 1, 2016 1:02 PM, doctorbulat <notifications@github.com> wrote:

what is the problem with input dynamic datasets? how to do it right?— Reply to this email directly or view it on GitHub.

doctorbulat commented 8 years ago

What about sequence? Can I use TWIST sequence (Siemens)? It is Time-resolved angiography With Interleaved Stochastic Trajectories. It is conventional 3D gradient echo sequence (“fl3d_ce”). In Siemens programm "Tissue 4D" they use TWIST sequence for DCE-MRI, but they investigate prostate cancer. May be you have some experience with 3T Siemens Verio !? And what about VIBE sequence (Siemens)? It is like fast SPGR on GE.

doctorbulat commented 8 years ago

Can I download all slices from the dynamic series (in 1 serie I have 20 slices for example) or just from 1 dynamic serie I can download 1 slice (for example 1 slice at the level of lateral ventricles) ?

doctorbulat commented 8 years ago

I made a mask in Fiji, but I cannot save it in .nii format. There are so many formats, but .nii is absent. So if my mask is not ok, can it be the reason of "input image sizes are not same'? The same question about ROI. Thank you!

ironictoo commented 8 years ago

TWIST sequence should work, VIBE should be ok also since it is just a spoiled gradient echo as far as I know. You can do an arbitrarily sized 4D dataset for the dynamic, no restrictions on slices. Although very large datasets with lots of slices and lots of time points take a very long time to process. Fiji should be able to save in .nii format, if it isn't working for you you'll have to look around on the FIJI website for help. Same with the ROI, you should should be able to save as *.roi file from within the ROI manager. I don't see how this would be related to "sizes not the same error", if your images are EXACTLY the same size and you still get this error let me know.

doctorbulat commented 8 years ago

Now everything is clear with runA part! Thank you!

Xueei commented 5 years ago

I'm new here.And I have trouble with runA part.Is directly click RUNA.fig and add dynamic datasets? Add test data which ROCKETSHIP provide? Why did I make the following mistakes?

尝试引用非结构体数组的字段。

出错 RUNA>addfiles_Callback (line 579) list = get(handles.DCEfilesA,'String');

出错 gui_mainfcn (line 95) feval(varargin{:});

出错 RUNA (line 42) gui_mainfcn(gui_State, varargin{:});

出错 @(hObject,eventdata)RUNA('addfiles_Callback',hObject,eventdata,guidata(hObject))

Error while evaluating UIControl Callback