open-discussion
open-discussion > RE: Convertion problems
Jun 8, 2017 01:06 PM | Martin Styner
RE: Convertion problems
Dear Ana
1. Yes, we know that DTIPrep does not load into the current Slicer's extension browser. One workaround is to use 4.4 like you did. We expect that DTIPrep will be available later this year in the current Slicer version. Having to keep up with Slicer versions is actually a clear downside of using Slicer as a dissemination infrastructure.
2. Not sure why you would use dicom to nrrd in your setting, if I understand correctly you want to convert your nifti (nii.gz) to nrrd . For that purpose use the same tool (DWIConvert) but you would need the newest Slicer version (the older DWIConvert had bugs in the nifti conversion)
3. Most importantly, I am not sure DTIPrep is the right tool for you. DTIPrep really designed for processing the DWI data (by getting rid of bad DWIs) not the fitted tensors. Btw, eddy current correction in DTIPrep is slightly better than FSL (though both are very similar) as it uses an iteratively averaged B0 for the correction. Now, if you would like to detect outliers in the fitted tensors there are other packages such as tortoise which provide an implementation of RESTORE for exactly this purpose.
Best regards
Martin
Originally posted by afouto:
1. Yes, we know that DTIPrep does not load into the current Slicer's extension browser. One workaround is to use 4.4 like you did. We expect that DTIPrep will be available later this year in the current Slicer version. Having to keep up with Slicer versions is actually a clear downside of using Slicer as a dissemination infrastructure.
2. Not sure why you would use dicom to nrrd in your setting, if I understand correctly you want to convert your nifti (nii.gz) to nrrd . For that purpose use the same tool (DWIConvert) but you would need the newest Slicer version (the older DWIConvert had bugs in the nifti conversion)
3. Most importantly, I am not sure DTIPrep is the right tool for you. DTIPrep really designed for processing the DWI data (by getting rid of bad DWIs) not the fitted tensors. Btw, eddy current correction in DTIPrep is slightly better than FSL (though both are very similar) as it uses an iteratively averaged B0 for the correction. Now, if you would like to detect outliers in the fitted tensors there are other packages such as tortoise which provide an implementation of RESTORE for exactly this purpose.
Best regards
Martin
Originally posted by afouto:
Hi,
I did some preprocessing steps on my diffusion data using FSL. More specifically, eddy-current distortions were corrected using FSL's eddy and then tensor fitting was performed using FSL's dtifit. At this stage I would like to perform an outliers detection with DTIPrep before starting Glm's analysis. However, I'm having trouble with the first stage of the convertion to Nrrd. I'm using Slicer 4.4 (linux version) instead of the most recent version because the extension for DTIPrep is not avaliable on the latter one. I followed the steps indicated in this tutorial to convert Dicom to Nrrd but nothing happends and I didn't even receive any error message (http://dmri.slicer.org/docs/tutorials/DW...). I also tried the convertion FSL to Nrrd but I didn't get any results either. The folder that I created to save the outputs remained empty. I'm more interested in the convertion FSL to Nrrd, once I already have the images preprocessed in the Nifti format.
I would appreciate some help to get over this first step and perform the outliers analyses.
Thanks in advance for the help
Best regards,
Ana F.
I did some preprocessing steps on my diffusion data using FSL. More specifically, eddy-current distortions were corrected using FSL's eddy and then tensor fitting was performed using FSL's dtifit. At this stage I would like to perform an outliers detection with DTIPrep before starting Glm's analysis. However, I'm having trouble with the first stage of the convertion to Nrrd. I'm using Slicer 4.4 (linux version) instead of the most recent version because the extension for DTIPrep is not avaliable on the latter one. I followed the steps indicated in this tutorial to convert Dicom to Nrrd but nothing happends and I didn't even receive any error message (http://dmri.slicer.org/docs/tutorials/DW...). I also tried the convertion FSL to Nrrd but I didn't get any results either. The folder that I created to save the outputs remained empty. I'm more interested in the convertion FSL to Nrrd, once I already have the images preprocessed in the Nifti format.
I would appreciate some help to get over this first step and perform the outliers analyses.
Thanks in advance for the help
Best regards,
Ana F.
Threaded View
| Title | Author | Date |
|---|---|---|
| afouto | Jun 7, 2017 | |
| Martin Styner | Jun 8, 2017 | |
| XIAO YANG | Jan 24, 2018 | |
| XIAO YANG | Jan 24, 2018 | |
| Martin Styner | Jan 26, 2018 | |
| Martin Styner | Jan 25, 2018 | |
| afouto | Jun 9, 2017 | |