NITRC Wisconsin White Matter Hyperintensities Segmentation Toolbox Forum: help

Error running code: not enough arguments in Whyd_setup.m

Hedaya Badr — Mon, 17 Jun 2024 19:37:48 GMT

Hi, 
I'm having issues running the code to completion to get the segment pmap and quanitification. Preprocessing and slicing runs fine through SPM12 but then stops when it gets to the probability and quantification portion of the code. Error: not enough input arguments. Error in width. Error in WhyD_gui, and the rest of the codes in the folder. 
Has anyone run into this issue and how do I fix it? 
 
Thanks

Compiling Requirements and Version Suggestions

jmatthews — Mon, 10 Feb 2020 20:50:17 GMT

Hello,

We are attempting to run this toolbox with Matlab R2019b on a CentOS6 server. When we try to run the install script or compile RegionGrowing_mex.cpp ourselves, we get an error stating our g++ version 4.4.7-17 is not supported and that we need version 6.3.x.

We were also confused because the toolbox will run without successfully running the install script or compiling anything, although our resulting prob maps are empty for an input with obvious lesions. Is this expected? I would have expected the toolbox to throw an error without the required script? Or is the toolbox intended to run with the included RegionGrowing_mex.m script?

Is there a version of matlab you would recommend on which the compilation is usually smooth?

Thank you

WMH lesion location in MNI space

Miguel Angel Rivas Fernández — Wed, 06 Feb 2019 22:47:04 GMT

Dear W2MHS developers,

I would like to use this interesting toolbox in order to detect and quantify WMH in FLAIR images. However, I would be specially interested in know and automatically extract the x,y,z MNI coordinates of the lesions detected in each subject.

[b]Is possible extract the MNI coordinates of lesions detected in each subject?[/b]

In case that this is not possible with the W2MHS toolbox, could you recomment to me some other alternative?

Any help would be appreciated.

Best,

Miguel

RE: Registration?

Will Khan — Tue, 20 Nov 2018 23:35:37 GMT

Hi,

The outputs generated by WhyD_preproc.m do not appear to be in either the FLAIR or T1 space. Since the defaults do not save a transformation matrix we are unable to decipher what inputs generate this 'warped' space. Could you please advise where specifically in the script one would need to alter the default spm jobman settings to allow transforms to be saved so that we can warp the outputs back to the original native space, please?

Thanks very much,

Cheers,

Will

Training, again

Salma Bent Ridha — Tue, 06 Nov 2018 13:49:13 GMT

Dear W2MHS,

Thank you for shearing such a great tool ! I was wondering if there has been some progress on the topic of training the RF model using my own images ? There is an old topic from 2015 about that saying it does not work at that time. Is it still impossible to achieve through the toolbox ?

Thanks a lot !
Best regards,
Salma

RE: Visualizing the result

jeremy sim — Wed, 13 Jun 2018 6:07:59 GMT

Apologies, I replied to the wrong thread earlier

RE: Problem with image intensity?

jeremy sim — Wed, 13 Jun 2018 6:06:55 GMT

Hi Laura,

I happened to face the same issue right now and haven't solve the issue despite changes in cut values and trying different preproc bias values (60, 80, 100, 120,140). I was wondering if you managed to overcome this problem previously, and if so, may I ask how did you do it?

Best regards,
Jeremy

RE: Visualizing the result

jeremy sim — Wed, 13 Jun 2018 6:05:44 GMT

RE: Compiling problems

Vamsi Ithapu — Thu, 22 Mar 2018 15:03:28 GMT

Hello.

We haven't encountered this problem with earlier versions of Matlab/Ubuntu. Please check out with an earlier Matlab version if possible. The mex-compile of Matlab doc will provide more info.

RE: Compiling problems

ericb — Thu, 22 Mar 2018 14:58:29 GMT

Sorry for the 1-person conversation, however I should point out that the above solution did not work. The file compiles but errors occur.

However, I found a more promising solution, that compiles, runs and appears to produce images as expected.

When compiling, the following flag is required:

[color=#000000]mex -compatibleArrayDims RegionGrowing_mex.cpp[/color]