help > Frequently Asked Questions
Showing 1-1 of 1 posts
Display:
Results per page:
Oct 7, 2013  07:10 PM | Christopher Lindner
Frequently Asked Questions
Q: What is W2MHS?
A: W2MHS is an open source MatLab toolbox for segmenting and quantifying white matter hyperintensities in T1 and T2 MRI images developed at the University of Wisconsin in conjunction with the Wisconsin Alzheimer's Disease Research Center (WADRC). 

Q: How can I get W2MHS?
A: The toolbox is available for download here on NITRIC and on SourceForge as well. There are some libraries required required to use this version of W2MHS. Make sure you have a C compiler, MatLab and SPM12b on your computer. A license for MatLab can be acquired from mathworks.com  and SPM12b can be found at: http://www.fil.ion.ucl.ac.uk/spm/software/spm12/

Q: What operation systems are supported?
A: W2MHS has been tested on both Linux and Windows workstations. If you have trouble installing this package on any operating system you can always ask for help in the help forums available here on NITRC and SourceForge.

Q: How much scripting knowledge do I need to use W2MHS?
A: None! While W2MHS is very powerful in its command line interface, the toolbox does come with an easy to use GUI. It has all of the same functionality as the command line and also allows you to save your sessions and import batches of images!

Q: What new features are included in version 1.3?
A: We are continually trying to improve both our user experience and the actual performance and accuracy of W2MHS. In the previous version, many users were having trouble installing on Windows machines. This problem has been resolved by tweaking our compile settings to be invariant to Windows and Linux operating systems. Other users were also having trouble with certain images not returning any WMH detections. This issue has also been resolved. 

In addition to these bug fixes, W2MHS also has a cleaner and easier to use GUI. The GUI also has new batch import feature allowing you to import hundreds of images into a batch at a time. Similar to this we have also included a batch script version that automatically assesses and segments an entire directory of images (given the right format).  If you have any additional feature requests, please drop us a line in the discussion forums. 

Q: Can I use my own features to train the supervised segmentation model?
A: Yes, you can. Our methodology is described in the corresponding paper which is still awaiting peer review. Documentation as well as scripts to assist in creating your own features will be released in the future.

You may ask any additional questions below!