I'm trying to reconstruct DTI parameters from my diffusion-weighted dataset, acquired with a Bruker system (paravision 5). My method of getting the Bruker data to the program is as follows:

1) use the Bruker utility for converting images into Dicom format
2) open the Dicom files in Slicer (at this point the data ordering is mixed up)
3) save dicom files as nhdr format
4) manually modify nhdr file to reflect data structure, b-value and gradient directions

I am successful in loading the data in a correct format (the volumes module tells me that I now have 30 DWI components which I can cycle through with the slider). My next step is to try calculating the DTI dataset by using the DWI to DTI estimation module. I specify the input DWI volume, as well as the output DTI volume and output baseline volume. The module says that it runs without errors, but the output volumes are both blank. And the error log provides the following warnings:

Warning: In ........\Slicer-431-1\Base\Logic\vtkSlicerApplicationLogic.cxx, line 1107
vtkSlicerApplicationLogic (00000000045D42E0): Unable to delete temporary file C:/Users/AYUNG/AppData/Local/Temp/Slicer/EFC_vtkMRMLDiffusionTensorVolumeNodeB.nrrd

Warning: In ........\Slicer-431-1\Base\Logic\vtkSlicerApplicationLogic.cxx, line 1107
vtkSlicerApplicationLogic (00000000045D42E0): Unable to delete temporary file C:/Users/AYUNG/AppData/Local/Temp/Slicer/EFC_vtkMRMLScalarVolumeNodeB.nrrd

I have provided a dropbox link to the dataset and nhdr file for your review. The example DTI dataset of human brain seems to work well, and it looks like my nhdr file looks the same. The only thing I noticed was the range of pixel values in my dataset seem to be an order of magnitude larger than the example DTI dataset...