Correlational Tractography

[irem yeşiloğlu]

Hi Frank,

I have two methodological questions regarding correlational tractography.

1. My dataset includes diffusion MRI acquired using different diffusion protocols (single-shell b=1000, multi-shell b=1000/2000, and a small number of higher b-value acquisitions). All datasets were reconstructed using QSDR. Is it appropriate to combine these subjects in a single correlational tractography analysis?

2. Is it methodologically appropriate to use total lesion volume as the study variable in correlational tractography to identify white matter pathways associated with overall lesion volume?

Thank you very much for your time.

[Frank Yeh]

Hi Irem,

These are good methodological questions.

1. If all subjects are reconstructed using QSDR, the data are brought into a common template space, so in principle they can be included in one connectometry/correlational tractography analysis. However, different diffusion protocols may still introduce systematic differences in QA or other diffusion indices. I would recommend including acquisition protocol as a covariate if possible, or running a sensitivity analysis to confirm that the finding is not driven by one protocol group. If the protocols are very different or strongly confounded with the clinical variable, then separate analyses would be safer.

2. Total lesion volume can be used as the study variable if it is a continuous subject-level measure. The interpretation would be pathways whose local diffusion measures are associated with overall lesion burden. It does not necessarily mean that the identified tracts are directly inside the lesions; rather, they are white matter segments whose microstructural measurements covary with total lesion volume.

I would also suggest controlling for age, sex, and other relevant variables, and checking whether lesion volume is strongly skewed. If it is highly skewed, a log transformation or nonparametric/sensitivity analysis may be useful.

Best,
Frank

On Sat, 27 Jun 2026 12:12:26 -0700 (PDT), “irem yeşiloğlu” iremyes...@gmail.com wrote:

Hi Frank,

I have two methodological questions regarding correlational tractography.

1. My dataset includes diffusion MRI acquired using different diffusion protocols (single-shell b=1000, multi-shell b=1000/2000, and a small number of higher b-value acquisitions). All datasets were reconstructed using QSDR. Is it appropriate to combine these subjects in a single correlational tractography analysis?

2. Is it methodologically appropriate to use total lesion volume as the study variable in correlational tractography to identify white matter pathways associated with overall lesion volume?

Thank you very much for your time.

–

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 visit https://groups.google.com/d/msgid/dsi-studio/a9bd6f99-64bf-4fb8-87d8-3346a68e77dan%40googlegroups.com.