On Sat, Mar 05, 2011 at 09:18:31PM +0100, martin f krafft wrote: > also sprach Michael Hanke <[email protected]> [2011.03.05.2034 +0100]: > > Sounds like you want to convert it to NIfTI format. We have a number of > > packages in Debian that can do that (dicomnifti, mriconvert, dcm2nii, > > ...). > > I am trying. I have three DVDs with DICOM/ directories, all filled > with STD0000xx directories containing IMG000xxx files. > > If I run e.g. > > % dinifti -g STD00002 std00002.nifti > > I am told > > **** DICOMImage::GetACQNumber: Cannot get value of image number in image > STD00002/IMG00001
Without looking at it in detail it is hard to say. Although DICOM is one standard it also has many dialects (each vendors cooks its own soup), which might be an issue here. but read on... > and no output file is generated. On another directory, I get > > **** DICOMImage::GetRepetitonTime: Cannot get value of acquisition > repetition time in image STD00001/IMG00001 This is a non-issue. This message is only relevant for 4D-images (multiple 3D images collected in rapid succession -- often for functional MRI, i.e. watching the brain work). Your data is probably 3D only. If you don't want to spend much time figuring this out, I'd say: loop over all directories and see on how much of it works out-of-the-box and if that is enough for you. On the remaining data you can try another converter (e.g. dcm2niigui from the mricron package), or you can look at the failing DICOM image series themselves to see what you are missing and whether it is worth the effort). For the latter I had a good experience with aeskulap (from the package of the same name). BTW: Once in NIfTI format -- most capable viewers will allow you to interactively view them in 'lightbox mode" (i.e. like a radiologist), or as orthogonal slices, and/or volume-rendered 3D models. Getting the brain surface meshes is a little more work (and error-prone), but the 'caret' package offers a tool for that too. Michael -- Michael Hanke http://mih.voxindeserto.de -- To UNSUBSCRIBE, email to [email protected] with a subject of "unsubscribe". Trouble? Contact [email protected] Archive: http://lists.debian.org/20110305212251.GB26056@meiner
