3 run DTI with the same values but different directions and reverse acquisition

1

Hi,
I have a dataset of single-shell DWI recordings, with 3 runs (identical parameters; dir=22, dir=23, dir=24, b=1000) per participant. I am planning to calculate FA, MD, RD, AD metrics, and reconstruct tractograms. I am new to imaging and was wondering if I should use the 24 dir run for my analysis or combine the data.
Also, I don’t see any reverse acquisition but in DICOM info dim4 is two times of the final converted data. Could be that?
Thank you

2

Hi @Akram,

Unless you are interested in test-retest reliability (which for such short runs, I wouldn’t be), you should combine the data. You can use tools such QSIPrep to process your data, which will take care of merging them correctly.

You can tell which image is reverse phase encoded by looking for the PhaseEncodingDirection field in the associated JSON files.

Best,
Steven

3

Hi @Steven ,

Thanks for your response. I checked the PhaseEncodingDirection field and It was the same for all.

Best,
Akram

4

Hi @Akram,

It sounds like none of your image are reverse phase encoded then.

Best,
Steven

1 Like
5

Hi @Akram,

If your data are coming from different sessions, e.g.:

then you’d want to be careful combining them. That is, you usually wouldn’t want to concatenate DWI scans that were several days (or more) apart from eachother.

Best,
Steven

6

Hi @Steven,

Thank you for your attention and sorry for the misleading language. The scans happened in the same imaging sessions (same day).

Best,
Akram