TubeTK/TBI Data: Difference between revisions

From KitwarePublic
Jump to navigationJump to search
No edit summary
No edit summary
 
(5 intermediate revisions by 2 users not shown)
Line 12: Line 12:


* '''Gradient Recalled Echo (GRE) data set [http://www.indiana.edu/~mri/CE/slides/Gradient%20Echo%20Sequences%20and%20Applications%20090324.pdf] [http://www.mr-tip.com/serv1.php?type=db1&dbs=Gradient%20Recalled%20Echo%20Sequence]:''' These images are used to generate contrast (by enhancing at high magnetic fields (7 T and above)) in MRI of the human brain. The main advantage of this data set is that it allows to visualize biological structures within gray matter (GM) and white matter (WM) that are not usually available with conventional MRI (e.g., blood deoxyhemoglobin, tissue lipid, and non-heme iron content.) <!-- Such sources as blood deoxyhemoglobin, tissue lipid, and non-heme iron content (5), as well as water-protein exchange (7) have been proposed as possible origins of the MRI signal frequency shift responsible for this contrast. -->[http://www.pnas.org/content/early/2009/07/22/0904899106.full.pdf]
* '''Gradient Recalled Echo (GRE) data set [http://www.indiana.edu/~mri/CE/slides/Gradient%20Echo%20Sequences%20and%20Applications%20090324.pdf] [http://www.mr-tip.com/serv1.php?type=db1&dbs=Gradient%20Recalled%20Echo%20Sequence]:''' These images are used to generate contrast (by enhancing at high magnetic fields (7 T and above)) in MRI of the human brain. The main advantage of this data set is that it allows to visualize biological structures within gray matter (GM) and white matter (WM) that are not usually available with conventional MRI (e.g., blood deoxyhemoglobin, tissue lipid, and non-heme iron content.) <!-- Such sources as blood deoxyhemoglobin, tissue lipid, and non-heme iron content (5), as well as water-protein exchange (7) have been proposed as possible origins of the MRI signal frequency shift responsible for this contrast. -->[http://www.pnas.org/content/early/2009/07/22/0904899106.full.pdf]
Preliminary brain, ventricle, and pathology segmentations are provided based on the T1 images for both patients at both acute and chronic time points. The relevant files are:
* '''[Acute/Chronic]_T1_brain.nii:''' Skull-stripped version of T1 image.
* '''[Acute/Chronic]_T1_labels(2).mha:''' Label maps for pathology regions and ventricles.
* '''[Acute/Chronic]_T1_lesion.vtk:''' 3D mesh model of pathology segmentation.
* '''[Acute/Chronic]_T1_ventricles.vtk:''' 3D mesh model of ventricle segmentation.




Line 85: Line 91:


b) Include the following text (or something similar) in your acknowledgements XXXXXXXXXXXXXXXXXX
b) Include the following text (or something similar) in your acknowledgements XXXXXXXXXXXXXXXXXX
[[Category:TubeTK Data|TBI Data]]

Latest revision as of 18:35, 26 July 2013

Image Format

Images are presented in the nii format. Each image can be loaded by any program that supports nii format, including most ITK and VTK-based applications (e.g., Slicer).

There are two patient data sets. For each of them, there are datasets for 4 different modalities for both Acute and Chronic Stage. These modalities are:

  • T1-weighted MRI: These are standard basic scans that are used to differentiate fat from water. They show water darker and fat lighter.
  • T2-weighted MRI: These are another standard basic scans that are used to differentiate fat from water. The difference of this data set from T1 is that they show fat darker, and water lighter.
  • Fluid-attenuated inversion recovery (FLAIR) data set: It is an advanced magnetic resonance imaging sequence that reveals tissue T2 prolongation with cerebrospinal fluid suppression, allowing detection of superficial brain lesions [1].
  • Gradient Recalled Echo (GRE) data set [2] [3]: These images are used to generate contrast (by enhancing at high magnetic fields (7 T and above)) in MRI of the human brain. The main advantage of this data set is that it allows to visualize biological structures within gray matter (GM) and white matter (WM) that are not usually available with conventional MRI (e.g., blood deoxyhemoglobin, tissue lipid, and non-heme iron content.) [4]

Preliminary brain, ventricle, and pathology segmentations are provided based on the T1 images for both patients at both acute and chronic time points. The relevant files are:

  • [Acute/Chronic]_T1_brain.nii: Skull-stripped version of T1 image.
  • [Acute/Chronic]_T1_labels(2).mha: Label maps for pathology regions and ventricles.
  • [Acute/Chronic]_T1_lesion.vtk: 3D mesh model of pathology segmentation.
  • [Acute/Chronic]_T1_ventricles.vtk: 3D mesh model of ventricle segmentation.


Susceptibility-weighted imaging (SWI) and DTI data sets are not available at the moment.

Detailed information about each data sets are provided in the table belows:

Data Set for Patient A
Acute/Chronic Time Point Flair GRE T1 T2
Resolution 384x512x50 384x512x50 256x256x176 320x320x 50
Spacing 0.46875x0.46875x3.00000 0.46875x0.46875x3.00000 1x1x1 0.75x0.75x3.0
Origin 90,
-105.12446594,
-70.34745789
90,
-105.12446594,
-70.34745789
128,
-102.42373657,
-75.87287903
120,
-104.84321594,
-70.34745789


Data Set for Patient B
Acute Time Point Flair GRE T1 T2
Resolution 384x512x53 384x512x53 256x256x176 260x320x53
Spacing 0.46875x0.46875x3.00000 0.46875x0.46875x3.00000 1x1x1 0.75x0.75x3.0
Origin 84.55205536,
-105.60873413,
-70.13075256
84.55205536,
-105.60873413,
-70.13075256
128,
-106.418884,
-85.684021
92.05205536,
-105.32748413,
-70.13075256


Data Set for Patient B
Chronic Time Point Flair GRE T1 T2
Resolution 384x512x50 384x512x50 256x256x176 512x512x50
Spacing 0.46875x0.46875x3.0000033 0.46875x0.46875x3.0000033 1x1x1 0.4687500x0.4687500x3.0000033
Origin 96.05750275,
-80.31092072,
-43.31170654
96.05750275,
-80.31092072,
-43.31170654
133.73010254,
-72.58300781,
-70.52586365
126.04978180,
-80.41034698,
-43.98503113

Terms of Use

This database is provided without charge for use in research, teaching, and commercial developments that advance the medical field. Uses outside of the medical field require prior written approval from Kitware. These data may not be redistributed except when a few cases are distributed as part of an electronic publication in which the author of the publication has added significant value to the publication, beyond the value of the cases being redistributed. In all other situations, citations and links to this webpage should be used.

We request that any publication or project that uses this data do the following:

a) Cite the following paper XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

b) Include the following text (or something similar) in your acknowledgements XXXXXXXXXXXXXXXXXX