Difference between revisions of "DTI Tractography Challenge Datasets"

From NAMIC Wiki
Jump to: navigation, search
Line 44: Line 44:
 
* 25 gradient directions + 1 baseline
 
* 25 gradient directions + 1 baseline
 
* Gradient directions were obtained using the b-matrix field of the dicom header
 
* Gradient directions were obtained using the b-matrix field of the dicom header
* Variable b-value (range between 50 and 1000): each diffusion weighted image has a different b value.
+
* Variable b-value (range between 40 and 1000): each diffusion weighted image has a different b value.
 
* Spacial resolution of 1.97x1.97x2mm
 
* Spacial resolution of 1.97x1.97x2mm
 
* Image size 96x96x81
 
* Image size 96x96x81

Revision as of 22:52, 11 May 2011

Home < DTI Tractography Challenge Datasets

Neurosurgical Datasets

Neurosurgical Case 1

  • Clinical Case: Astrocytoma, World Health Organization (WHO) grade III
  • Acquistion Parameters

Data were acquired on a 3.0-T scanner (EXCITE Signa scanner, GE Medical System, Milwaukee, Wisconsin) with Excite 14.0 using an 8-channel head coil.
T1 SPGR: TR=7500 ms; TE=30 ms; matrix 256x256; FOV 25.6 cm ; 1-mm slice thickness
T2 GRE: TR=8000 ms; TE=98 ms; matrix 512x512; voxel size 0.5 x 0.5 x 1.5 mm
DWI Singe shot EPI: TR=14 000ms; TE=30 ms; 31 gradient directions, 1 baseline ; b-value = 1000 s/mm2 ; matrix 128x128; 2.6 mm slice thickness
Segmentation: The file patient1-tumor.nhdr & raw is labelmap which contains the segmented structures: label #1 represents the solid part of the tumor; label #2 represents the cystic part of the tumor.

Neurosurgical Case 2

  • Clinical Case: Oligoastrocytoma,WHO grade II
  • Acquistion Parameters

Data were acquired on a 3.0-T scanner (EXCITE Signa scanner, GE Medical System, Milwaukee, Wisconsin) with Excite 14.0 using an 8-channel head coil.
T1 SPGR: TR=7500 ms; TE=30 ms; matrix 256x256; FOV 25.6 cm ; 1-mm slice thickness
T2 GRE: TR=8000 ms; TE=98 ms; matrix 512x512; voxel size 0.5 x 0.5 x 1.5 mm
DWI Single shot EPI: TR=14 000ms; TE=30 ms; 55 gradient directions, 1 baseline ; b-value = 1000 s/mm2 ; matrix 128x128; 2.6 mm slice thickness
Segmentation: The file patient2-tumor.nhdr & raw is labelmap which contains the segmented structures: label #1 represents the solid part of the tumor; label #2 represents the cystic part of the tumor.

Neurosurgical Case 3

  • Clinical Case: Pending
  • Acquisition Parameters:
  • T1 Axial SPGR: matrix 256x256; voxel size 1 x 1 x 1.4 mm, volume size 256 x 256 x130
  • T2 matrix 512x512; voxel size 0.5 x 0.5 x 1.5 mm, volume size 512 x 512 x 100
  • DWI matrix 256 x256; voxel size 1 x 1 x 2.6 mm, 31 gradient directions, 1 baseline , b-value = 1000 s/mm2; volume size 256 x 256 x 32 x 52


Neurosurgical Case 4

  • Clinical Case: Pending
  • Acquisition Parameters:
  • T1 Axial SPGR: matrix 256x256; voxel size 1 x 1 x 1.4 mm, volume size 256 x 256 x130
  • T2 matrix 512x512; voxel size 0.5 x 0.5 x 1.5 mm, volume size 512 x 512 x 100
  • DWI matrix 256 x256; voxel size 1 x 1 x 2.6 mm, 31 gradient directions, 1 baseline , b-value = 1000 s/mm2; volume size 256 x 256 x 32 x 52

Control subjects Datasets

Two adult control subject datasets will be available for this challenge. For each subject we have 10 repetitions of the structural and diffusion-weighted acquisitions. All scans were acquired on the same type of scanner (Siemens Tim Trio) at 5 different institutions.

Details of the DWI acquisition parameters:

  • 25 gradient directions + 1 baseline
  • Gradient directions were obtained using the b-matrix field of the dicom header
  • Variable b-value (range between 40 and 1000): each diffusion weighted image has a different b value.
  • Spacial resolution of 1.97x1.97x2mm
  • Image size 96x96x81

Details of DTI estimation:

  • Tensor estimation using weighted least square algorithm

Details of the structural acquisition. We provide two sets of images:

  • Original T1 and T2 (size: 160x224x256, spacing: 1x1x1mm)
  • T1 and T2 registered and resampled to baseline (size: 96x96x81, spacing: 1.97x1.97x2mm) (Affine + BSpline)

File format and naming convention for submission

For each case, the results should be submitted in the following formats: 1) 3D coordinate of the tracts: vtkPolydata ASCII VTK file format; 2) enveloppe of the tracts: ITK-readable Nrrd file format
Naming convention:

  • zip archive file: miccaiDTIChallenge_firstAuthorName.zip
  • VTK file: firstAuthorName_tract_coordinate.vtk
  • Nrrd file: firstAuthorName_tract_enveloppe.nhdr and firstAuthorName_tract_enveloppe.nhdr
  • PNG file: firstAuthorName_tract_axialView.png; firstAuthorName_tract_sagittalView.png; firstAuthorName_tract_coronalView.png



Back to DTI Tractography Challenge MICCAI 2011