Ir para o conteúdo

 Voltar a invesalius-int
Tela cheia

Strange ordering of DICOM images once imported

11 de Abril de 2011, 21:30 , por Desconhecido - | Ninguém seguindo este artigo por enquanto.
Visualizado 210 vezes

Hi, all. I'm a new Invesalius user and not experienced with medical imaging. Can anyone suggest what I'm doing wrong, or what else I can do when configuring the import, given the following bug/feature I experience.

The ordering shown in the 'tree view' when I first select to import the images is correct (the slice goes from smaller sections through larger sections and then smaller as you'd expect from an ordered series), but the order shown in the bottom left 'coronal slice' in the full project view after the import has completed is totally wrong, and the 3d model corresponds with this confused ordering; very little sensible structure can be extracted. 

I'm trying to solve a problem for a friend who's not hyper-technical and has some images from a recent scan but wants a 3d model instead, though this is just for fun, rather than for serious diagnostics. 

I've recieved a series of DICOM images which seem properly labelled, quite consistent and well ordered when viewed in a directory by name, or through the import wizard of Invesalius. However, when the import has completed, the sequence seems to be randomised, with individual slices no longer in order. Consequently the 3D model is completely junked. 

Invesalius has been installed with the latest Sigar from git and dependencies satisfied from Ubuntu Lucid or Debian Sid repositories, and has stopped complaining about missing python modules, so I don't think there's anything wrong with the build.

Suggestions welcomed. 

Autor: Cefn Hoile


22 comentários

  • C7a30d8d4b033c119131929b2ec1d214?only path=false&size=50&d=404Cefn Hoile(usuário não autenticado)
    11 de Abril de 2011, 22:00

     

    Here's a dump of the embedded information inside an individual DICOM image from the sample set, as reported by Imagemagick's identify command, (with personal information stripped out), in case it helps... cefn@cefn-ubuntu-dell:~/Documents/code/neckslices$ identify -verbose 92221665 Image: 92221665  Format: DCM (Digital Imaging and Communications in Medicine image)  Class: PseudoClass  Geometry: 208x256+0+0  Resolution: 72x72  Print size: 2.88889x3.55556  Units: Undefined  Type: Grayscale  Base type: Grayscale  Endianess: LSB  Colorspace: RGB  Depth: 12/16-bit  Channel depth:    gray: 12-bit  Channel statistics:    gray:      min: 433 (0.105745)      max: 4095 (1)      mean: 750.207 (0.183159)      standard deviation: 591.303 (0.144363)      kurtosis: 11.1632      skewness: 3.18567  Colormap: 65536  Rendering intent: Undefined  Interlace: None  Background color: white  Border color: rgb(223,223,223)  Matte color: grey74  Transparent color: black  Compose: Over  Page geometry: 208x256+0+0  Dispose: Undefined  Iterations: 0  Compression: Undefined  Orientation: Undefined  Properties:    date:create: 2011-04-04T15:20:58+01:00    date:modify: 2011-04-04T15:19:26+01:00    dcm:: A1/PFP/SAT2     dcm:?: SIEMENS MEDCOM HEADER2    dcm:AcquisitionDate: 20090612    dcm:AcquisitionNumber: 1     dcm:AcquisitionTime: 110734.692500     dcm:AngioFlag: N     dcm:dB/dt: 0     dcm:DeviceSerialNumber: 25691     dcm:EchoNumber(s): 1     dcm:EchoTime: 86    dcm:EchoTrainLength: 256     dcm:FlipAngle: 150     dcm:FrameofReferenceUID: 1.3.12.2.1107.5.2.30.25691.2.20090612105609421.0.0.0    dcm:ImageDate: 20090612    dcm:ImagedNucleus: 1H    dcm:ImageOrientation(Patient): 1\0\0\0\0.01570730921325\-0.9998766326089     dcm:ImagePosition(Patient): -160.25007963181\-96.4462180072\173.14691518106     dcm:ImageText: SIEMENS MR HEADER     dcm:ImageTime: 110806.562000     dcm:ImageType: ORIGINAL\PRIMARY\M\NORM\DIS2D     dcm:ImagingFrequency: 63.69214    dcm:ImplementationClassUID: 1.3.12.2.1107.5.2    dcm:ImplementationVersionName: MR_VB13A    dcm:Instance(formerlyImage)Number: 1     dcm:InstanceCreationDate: 20090612    dcm:InstanceCreationTime: 110806.562000     dcm:InstitutionAddress:     dcm:InstitutionName:      dcm:MagneticFieldStrength: 1.5     dcm:Manufacturer: SIEMENS     dcm:Manufacturer'sModelName: Avanto    dcm:MediaStorageSOPClassUID: 1.2.840.10008.5.1.4.1.1.4    dcm:MediaStorageSOPInstanceUID: 1.3.12.2.1107.5.2.30.25691.3.2009061211073670727114644    dcm:Modality: MR    dcm:MRAcquisitionType: 2D    dcm:NumberofAverages: 1     dcm:NumberofPhaseEncodingSteps: 100     dcm:Patient'sAge:     dcm:Patient'sBirthDate:     dcm:Patient'sID:      dcm:Patient'sName:     dcm:Patient'sSex:      dcm:Patient'sWeight:     dcm:PatientPosition: HFS     dcm:PercentPhaseFieldofView: 81.25     dcm:PercentSampling: 75.9615     dcm:PhaseEncodingDirection: ROW     dcm:PhotometricInterpretation: MONOCHROME2     dcm:PixelBandwidth: 390     dcm:PixelSpacing: 1.5625\1.5625     dcm:ProtocolName: t2_haste_cor_bh_new     dcm:ReferencedImageSequence: ��    dcm:RepetitionTime: 1000    dcm:RequestedProcedureDescription: NIKA C-spine    dcm:SAR: 1.0108323748789     dcm:ScanningSequence: SE    dcm:ScanOptions: PFP\SAT2    dcm:SequenceName: *h2d1_158     dcm:SequenceVariant: SK\SP\OSP     dcm:SeriesDate: 20090612    dcm:SeriesDescription: t2_haste_cor_bh_new     dcm:SeriesInstanceUID: 1.3.12.2.1107.5.2.30.25691.3.2009061211073631182514636.0.0.0    dcm:SeriesNumber: 6     dcm:SeriesTime: 110806.546000     dcm:SliceLocation: -93.714647552833    dcm:SliceThickness: 4     dcm:SoftwareVersion(s): syngo MR B13 4VB13A     dcm:SOPClassUID: 1.2.840.10008.5.1.4.1.1.4    dcm:SOPInstanceUID: 1.3.12.2.1107.5.2.30.25691.3.2009061211073670727114644    dcm:SpacingBetweenSlices: 5.2     dcm:SpecificCharacterSet: ISO_IR 100    dcm:StationName: MRC25691    dcm:StorageMediaFileSetUID: 1.3.12.2.1107.5.2.30.25691.30000009061204312300000000032    dcm:StudyDate: 20090612    dcm:StudyDescription: NIKA^C-spine    dcm:StudyID: 1     dcm:StudyInstanceUID: 1.3.12.2.1107.5.2.30.25691.30000009061204044860900000010    dcm:StudyTime: 105609.250000     dcm:TransferSyntaxUID: 1.2.840.10008.1.2.1    dcm:TransmittingCoil: Body    dcm:VariableFlipAngleFlag: N     dcm:WindowCenter: 178     dcm:WindowCenter&WidthExplanation: Algo1     dcm:WindowWidth: 454     signature: 9eeb2678e03d179a6185791893e53a1c80ceb26307c627fb172603d9af6a13a0  Artifacts:    verbose: true  Tainted: False  Filesize: 183KiB  Number pixels: 52KiB  Pixels per second: 5.078MiB  User time: 0.010u  Elapsed time: 0:01.010  Version: ImageMagick 6.5.7-8 2010-12-02 Q16 www.​imag​emag​ick.​org

    • 4fc11d72fb1974e943174c8ca2fa8765?only path=false&size=50&d=404Thiago Franco Moraes(usuário não autenticado)
      12 de Abril de 2011, 9:07

       

      Hi Cefn Hoile,

      It's very strange. Are you using which version of InVesalius? Could you send us those dicoms?

      I didn't know imagemagick shows informations from dicom files, good to know that. thanks!

      Thanks!

Tutorial passo-a-passo sobre uso do software InVesalius 3

17 de Abril de 2010, 11:45, por Desconhecido

Se você tem interesse em utilizar o software de reconstrução 3D de imagens médicas InVesalius, mas não sabe por onde começar, acesse já o tutorial escrito pelo designer Cícero Moraes:



InVesalius 3 Beta 2 disponível

6 de Março de 2010, 11:41, por Desconhecido

Para saber mais sobre o InVesalius 3.0.0 Beta 2, leia:
http://svn.softwarepublico.gov.br/trac/invesalius/wiki/releases/pt/changelog

Continue contribuindo com a Comunidade InVesalius!

------------------------------------



Siga cada passo do InVesalius com o Twitter!

8 de Fevereiro de 2010, 9:57, por Desconhecido

Acompanhar o InVesalius pelo Twitter permitirá que você saiba, em primeira mão, sobre:



InVesalius 3 Beta 1 disponível para testes em Windows e GNU Linux

27 de Janeiro de 2010, 10:50, por Desconhecido



Trabalho acadêmico discute aplicação do OpenBRR ao InVesalius

6 de Novembro de 2009, 16:35, por Desconhecido

Autor: Tatiana Al-Chueyr Pereira Martins