Wrong orientations after loading src and fib files

[Shihan]

Hi Frank,

Thank you for developing the great software!

Currently I'm trying to track muscle fibres using the DTI data from human forearm. It's quite different from processing brain DTI data. I'm new to this field and encountered with some problems while using DSI studio. Hope that you can give me some advice. Thank you.

A file named 'Mosaic_se15_data.zip' has been uploaded. DTI data are named with a prefix - 'se15'. T1W data in NIfTI format is named 't1w'. There is also a mask file generated using DSI studio during step 2.

1. When I loaded the fib data to track fibers, I found that the orientation was wrong. I clicked the "Switch to axial view" in the left bottom of the GUI. However, it still showed the sagittal view. The axial slice was shown at the coronal view. T1W data can be inserted and its orientation was correct.

2. During step T2: Reconstruction, a mask should be generated to cover white matter. Should I adjust the mask to cover all muscles of interest? Or could I just dilate the mask to cover the whole field?

3. Do you think it is necessary to correct eddy current and motion distortion before calculating tensors for limb DTI data? I tried FSL 'eddy' or 'eddy_correct' function to account for these artefacts. However, the resultant NIfTI image seems to be equivalent with the original one. Therefore, I only performed Rician denoising to the original NIfTI data.

Do you have any ideas of the reasons for the wrong orientations? And could you please give me some suggestions on fiber tracking for limb DTI data? I'm using the DSI studio downloaded on March 15, 2021.

Thank you!

Best regards,

Shihan

[Frank Yeh]

>
> 1. When I loaded the fib data to track fibers, I found that the orientation was wrong. I clicked the "Switch to axial view" in the left bottom of the GUI. However, it still showed the sagittal view. The axial slice was shown at the coronal view. T1W data can be inserted and its orientation was correct.

It seems that the nifti file from FSL is causing the orientation
problem because this is not brain images. The bval and bvec are not
correctly oriented.

My suggestion is to use DSI Studio to parse DICOM images and create SRC files.

SRC files can be converted to NIFTI with bval and bvec in [Step
T2][File] menu for FSL processing.

The processed NIFTI files can be used to create another SRC file.

Once you create an SRC file from DICOM, you can send it to me, and I
can help check it again.

>
> 2. During step T2: Reconstruction, a mask should be generated to cover white matter. Should I adjust the mask to cover all muscles of interest? Or could I just dilate the mask to cover the whole field?

For DTI analysis, a precise mask can avoid background noise in FA.
For GQI analysis, it does not matter, and you can use threshold=0 to
include all image volume.

>
> 3. Do you think it is necessary to correct eddy current and motion distortion before calculating tensors for limb DTI data? I tried FSL 'eddy' or 'eddy_correct' function to account for these artefacts. However, the resultant NIfTI image seems to be equivalent with the original one. Therefore, I only performed Rician denoising to the original NIfTI data.

For publication purposes, yes, even if it does not make much of the differences.

Best regards,
Frank
>
> --
> You received this message because you are subscribed to the Google Groups "DSI Studio" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to dsi-studio+...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/dsi-studio/86b9a697-73ff-4a06-8530-db6566fe588co%40googlegroups.com.

[Shihan]

Hi Frank,

Thanks for your reply! 

I used the raw DICOM files to create SRC file and the FIB file. It seems that the slice orientations are correct this time. A file named 'se15_data_revised.zip' has been uploaded, in which the 'log' file describes detailed processing steps.

By the way, I wondered whether the T1W images and the DTI data are automatically aligned when inserting T1W images during Step T3. I see that there is an 'Adjust Registration' option in [Slice].

Best regards,

Shihan

> To unsubscribe from this group and stop receiving emails from it, send an email to dsi-s...@googlegroups.com.

[Frank Yeh]

DSI Studio will register them together for you.
[Adjust Registration] is for conditions when this registration does
not work out well. Users can manually align them together.
Frank

> To unsubscribe from this group and stop receiving emails from it, send an email to dsi-studio+...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/dsi-studio/50d35cbe-92b4-44dd-9870-a1f2f06252c1o%40googlegroups.com.

[Frank Yeh]

I checked the updated SRC files. It should work out for you.
Frank

[Shihan]

Hi Frank,

Thanks for your answers. It really helps.

Best regards,

Shihan

> > To view this discussion on the web visit https://groups.google.com/d/msgid/dsi-studio/50d35cbe-92b4-44dd-9870-a1f2f06252c1o%40googlegroups.com.