gbarisano
commented
3 days ago Hi Marcel,
Thank you for your message! And sorry for the delay in my response. Here my answers to your questions: 1) No, I did not design the script to specifically process multi-shell DTI data. The tensor model which is fitted by my script is generally not a good model for multi-shell data: as the b-value increases, the diffusion signal becomes sensitive to more complex microstructural features of tissue, such as cell membranes, myelin, and other barriers that restrict or hinder water diffusion. These structures lead to non-Gaussian diffusion patterns, which the tensor model cannot capture accurately. The DTI model typically works with single-shell data, having b0 value and typically b-value = 1000s/mm^2 (maximum b-value 1200s/mm^2, less frequently up to 1500s/mm^2) 2) Despite the point above, even though by a theoretical point of view it might seem inappropriate to calculate DTI-ALPS on multi-shell data with multiple b-values above 1200s/mm^2, I did some tests on multi-shell data with my script and the ALPS results were reasonable. I also tried to re-run the script using only the b0 images and the b values close to 1000 or below 1000, but the DTI model did not fit very well. One thing to keep in mind is that FSL's eddy in my script likely will not work for multi-shell data. Based on these tests, I think that as long as the FA maps look good, the ALPS results should be acceptable, but the actual values likely will not be directly comparable with DTI-ALPS values obtained from traditional single-shell DTI, so you would need to make sure that all the diffusion data included in your study have been acquired with the same multi-shell protocol. I will try to run some additional tests and assess more systematically DTI-ALPS on multi-shell data.
I hope this helps. Let me know if you have any question or suggestion.
Giuseppe
Dear Dr. Barisano,
Thank you very much for sharing this great code. I am still new to the topic of DTI ALPS. In the 2024 paper "Diffusion Tensor Image Analysis Along the Perivascular Space (DTI-ALPS): Revisiting the Meaning and Significance of the Method" by Taoka et al., it is stated that "The use of a single b-value instead of multiple shells makes it impossible to separate the velocity components of diffusion, and the information is considered to be a mixture of diffusion components of different velocities. This issue needs to be verified in the future."
As part of a study, we acquired a multiple shell DTI sequence with various b values (5 2005 995 2000 1995 995 2000 2005 1000).
1.I successfully processed a subject yesterday, and the ALPS indices are about 1.5 on both the right and left sides. I wanted to ask if the script is designed to also process multiple-shell DTI, or would I need to make some adjustments?
Best regards, Marcel