help
60 Subscribers 
help > Phillips DICOM to .nii conversion error
Showing 1-3 of 3 posts
Jun 9, 2021 02:06 PM | Tyler Diorio - Purdue University, Weldon School of Biomedical Engineering
Phillips DICOM to .nii conversion error
Hello,
I'm working with a set of 4D Flow MRI data acquired on a Phillips scanner. When attempting to convert the single folder of DICOM images using the MRIcroGL dcm2niix package I'm met with the following message:
I've read the previous forum post (https://www.nitrc.org/forum/message.php?msg_id=27724), but I've verified that my data (~/dcm) is not contained in subfolders (~/dcm/a; ~/dicom/b; ~/dicom/c). Additionally, I don't have information about the MRI sequence being downloaded from the machine so I'm not able to verify that all the files were copied correctly or that the sequence was interrupted mid-volume.
However, I am wondering if there is any way for me to work through this error assuming that all the files were obtained correctly and the sequence was uninterrupted.
Thank you for your help!
Tyler
I'm working with a set of 4D Flow MRI data acquired on a Phillips scanner. When attempting to convert the single folder of DICOM images using the MRIcroGL dcm2niix package I'm met with the following message:
Warning: Unable to determine slice direction:
please check whether slices are flipped
Convert 1108 DICOM as C:\Users\YYY\Documents\XXX\4DFLOWanon_4D-TRANCE_20200729000000_701 (336x336x1108x1)
Saving as 32-bit float (slope, intercept or bits allocated varies).
Unable to equalize slice distances: slice order not consistently ascending: [dx = [0 0 0 ... 90.35]
Recompiling with '-DmyInstanceNumberOrderIsNotSpatial' might help.
Convert 1108 DICOM as C:\Users\YYY\Documents\XXX\4DFLOWanon_4D-TRANCE_20200729000000_701 (336x336x1108x1)
Saving as 32-bit float (slope, intercept or bits allocated varies).
Unable to equalize slice distances: slice order not consistently ascending: [dx = [0 0 0 ... 90.35]
Recompiling with '-DmyInstanceNumberOrderIsNotSpatial' might help.
I've read the previous forum post (https://www.nitrc.org/forum/message.php?msg_id=27724), but I've verified that my data (~/dcm) is not contained in subfolders (~/dcm/a; ~/dicom/b; ~/dicom/c). Additionally, I don't have information about the MRI sequence being downloaded from the machine so I'm not able to verify that all the files were copied correctly or that the sequence was interrupted mid-volume.
However, I am wondering if there is any way for me to work through this error assuming that all the files were obtained correctly and the sequence was uninterrupted.
Thank you for your help!
Tyler
Jun 9, 2021 02:06 PM | Chris Rorden
RE: Phillips DICOM to .nii conversion error
Tyler-
You want to inspect the first line of the dcm2niix output
Chris Rorden's dcm2niiX version v1.0.20210410 Clang12.0.0 ARM (64-bit MacOS)
and make sure you are using the latest stable release (v1.0.20210317)
https://github.com/rordenlab/dcm2niix/releases
I note that the sequence is called 4D but the software thinks there are only 3 dimensions (336x336x1108x1). My best guess would be that an anonymization stage removed some of the critical spatial meta data from this scan. Is the problem resolved if you export the data directly off the scanner?
You want to inspect the first line of the dcm2niix output
Chris Rorden's dcm2niiX version v1.0.20210410 Clang12.0.0 ARM (64-bit MacOS)
and make sure you are using the latest stable release (v1.0.20210317)
https://github.com/rordenlab/dcm2niix/releases
I note that the sequence is called 4D but the software thinks there are only 3 dimensions (336x336x1108x1). My best guess would be that an anonymization stage removed some of the critical spatial meta data from this scan. Is the problem resolved if you export the data directly off the scanner?
Jun 10, 2021 04:06 PM | Tyler Diorio - Purdue University, Weldon School of Biomedical Engineering
RE: Phillips DICOM to .nii conversion error
Chris,
Thank you for your reply! I think the issue may very well be in the anonymization process. I'm going to reach back out to our data provider and ensure that the spatial meta data hasn't been lost.
I'll be sure to post an update here when I have more information.
Thank you,
Tyler
Thank you for your reply! I think the issue may very well be in the anonymization process. I'm going to reach back out to our data provider and ensure that the spatial meta data hasn't been lost.
I'll be sure to post an update here when I have more information.
Thank you,
Tyler