DWIConvert Training Resources

From NAMIC Wiki
Revision as of 16:07, 18 June 2014 by Farukhk (talk | contribs)
Jump to: navigation, search
Home < DWIConvert Training Resources

This page is work in progress.

DWIConverter Documentation

Parameters of the Slicer4.3 version

  • Common Parameters: Parameters used in all modes
  • Input DWI volume file (inputVolume): Input DWI volume -- not used for DicomToNrrd mode.
  • Output Filename (outputVolume): Output filename (.nhdr or .nrrd)
  • Output fMRI file (fMRIOutput): Output a NRRD file, but without gradients
  • DicomToNrrd Parameters: Parameters for Dicom to NRRD Conversion
  • Input Dicom Data (inputDicomDirectory): Directory holding Dicom series
  • Output Directory (outputDirectory): Directory holding the output NRRD file
  • Gradient Vector File (gradientVectorFile): Text file giving gradient vectors
  • Small Gradient Threshold (smallGradientThreshold): If a gradient magnitude is greater than 0 and less than smallGradientThreshold, then *DWIConvert will display an error message and quit, unless the useBMatrixGradientDirections option is set.
  • Output Options: Options to control the output.
  • Write Protocol Gradients File (writeProtocolGradientsFile): Write the protocol gradients to a file suffixed by ".txt" as they were specified in the protocol by multiplying each diffusion gradient direction by the measurement frame. This file is for debugging purposes only, the format is not fixed, and will likely change as debugging of new dicom formats is necessary.
  • Use Identity Measurement Frame (useIdentityMeaseurementFrame): Adjust all the gradients so that the measurement frame is an identity matrix.
  • Use BMatrix Gradient Directions (useBMatrixGradientDirections): Fill the nhdr header with the gradient directions and bvalues computed out of the BMatrix. Only changes behavior for Siemens data. In some cases the standard public gradients are not properly computed. The gradients can emperically computed from the private BMatrix fields. In some cases the private BMatrix is consistent with the public grandients, but not in all cases, when it exists BMatrix is usually most robust.
  • FSLToNrrd Parameters: FSLToNrrd Parameters
  • B Values File (inputBValues): B Values text file
  • B Vector File (inputBVectors): B Vector text file
  • NrrdToFSL Parameters: NrrdToFSL Parameters
  • B Values File (outputBValues): B Values text file
  • B Vector File (outputBVectors): B Vector text file


‹writeProtocolGradientsFile ‹ writes the DWI gradient vectors as a simple text file, used for debugging.

‹useIdentityMeasurementFrame ‹ more complicated: There are two reference frames with respect to DWI file. There¹s the image reference frame, which defines the patient position and orientation for the gradient magnitude voxel grid, and there is a measurement frame, which is the reference frame of the gradient vectors. This flag modifies the gradients written out so that they are in the Image reference frame.

‹useBMatrixGradientDirections

 From DWIConvert ‹help output: Fill the nhdr header with the gradient

directions and bvalues computed out of the BMatrix. Only changes behavior for Siemens data. In some cases the standard public gradients are not properly computed. The gradients can emperically computed from the private BMatrix fields. In some cases the private BMatrix is consistent with the public grandients, but not in all cases, when it exists BMatrix is usually most robust.

In other words, the Siemens data includes both a Bmatrix tag, and an explicit set of gradient vectors. For some DWI datasets, the scanner puts incorrect data in the explicit gradients, but the correct gradients can be recovered by computations on the Bmatrix. This option selects the latter methd.

2. The manufacturers supported are

 Siemens, Philips, GE and Hitachi. The Hitachi support is experimental,

as it has only been tested with scans from one site. Remove any reference to GDCM.

3. ‹fMRIOutput ‹ this outputs a NRRD file, but doesn¹t include gradient vectors. It would be used if you were interested in converting functional MRI datasets, instead of DWI datasets.