siemens dicom reading error

[Satrajit Ghosh]

None of the dicom files in the following archive are treated as valid. i can see the data and tags when i set force=True in read_file.

http://dl.dropbox.com/u/363467/dicom.tar.gz

phantom data using the 32Ch coil. A standard protocol of low resolution EPI was used, with TR=2, TE=30, 10 time-points(measurements), no partial fourier, resolution=3.1x3.1x4, 32 slices full head coverage, AC-PC orientation, interleaved slice acquisition,  20% slice gap and NO PACE. Phantom was the standard Siemens sphere. 2 sequences were used reflected at the data name:
ge_functionals_MITMGH for the custom sequence and
ep2d_bold_ProductSiemens for the Siemens product ep2d

i've also posted this on the nipy list for the nibabel folks to take a look.

[Satrajit Ghosh]

just checking to see, if anybody has been able to take a look at what's causing the exception.

[Darcy Mason]

On Jul 27, 3:36 pm, Satrajit Ghosh <satrajit.gh...@gmail.com> wrote:
> None of the dicom files in the following archive are treated as valid. i can
> see the data and tags when i set force=True in read_file.

On Aug 21, 11:33 pm, Satrajit Ghosh <satrajit.gh...@gmail.com> wrote:
> just checking to see, if anybody has been able to take a look at what's
> causing the exception.

Apologies, I always try to respond if no one else has, but somehow
missed this while away on business.

I took a quick look at a couple of files using the pydicom dump
utility; they did not have the required DICOM preamble and 'DICM'
marker for media storage. Unfortunately some systems export files that
way, which is why the force=True option exists. There may be a config
option on your system which can fix that, or you will just have to
always use the force option.

Regards,
Darcy

[Satrajit Ghosh]

hi,

i received the following error when reading two sets of dicom files. i did set force=True, but that still resulted in an exception. is there a way to bypass the exception?

code:

tf = tarfile.open(tarfilename)

tfi = tf.getmembers()

fp = tf.extractfile(tfi[0])

ds = dicom.read_file(fp, force=True)

errors:

---

Could not read /software/temp/dicom_metadata/data/PhilipsAchieva6.tar.gz:

err: 'Tag (2001, 105f) not found in DICOM dictionary' Could not read /software/temp/dicom_metadata/data/PhilipsAchieva7.tar.gz: err: 'Tag (2001, 105f) not found in DICOM dictionary'

---

source dicoms:

PhilipsAchieva6.tar.gz',
  'http://midas.kitware.com/bitstream/view/7442'),
PhilipsAchieva7.tar.gz',
  'http://midas.kitware.com/bitstream/view/7443')

cheers,

satra

[Darcy Mason]

On Oct 14, 1:12 pm, Satrajit Ghosh <sa...@mit.edu> wrote:
> ...

> tf = tarfile.open(tarfilename)
> tfi = tf.getmembers()
> fp = tf.extractfile(tfi[0])
> ds = dicom.read_file(fp, force=True)
>
> errors:
>
> ---
> Could not read /software/temp/dicom_metadata/data/PhilipsAchieva6.tar.gz:
> err: 'Tag (2001, 105f) not found in DICOM dictionary' Could not read

> ...

Hi Satra,

It seems this is fixed in the latest revision (worked for me, and
force=True not needed). Probably related to issue 97 [1], which has
been fixed.

Cheers,
Darcy

[1] http://code.google.com/p/pydicom/issues/detail?id=97

[Satrajit Ghosh]

thanks! that worked.

cheers,

satra

--
You received this message because you are subscribed to the Google Groups "pydicom" group.
To post to this group, send email to pyd...@googlegroups.com.
To unsubscribe from this group, send email to pydicom+u...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/pydicom?hl=en.