search

QMICodeBase / TORTOISEV4

Official TORTOISE Diffusion MRI Processing Pipeline V4 Source Code and Documentation
GNU General Public License v3.0
18 stars 4 forks source link

B value changes in DRBUDDI, should I expect the same behavior here in V4 #12

Open johnccshen opened 5 months ago

johnccshen commented 5 months ago

Hi team,

Thanks for providing the great tool for the community to use. I am new user of TORTOISE.

I noticed that when using DRBUDDI in TORTOISE V3.2 the b value changes from 1000 to 99x. I would like to know the reason behind.

If I switch the to use V4. Should I expect the same behavior?

Thanks in advanced!

johnccshen commented 5 months ago

Realized that the differences might came from TORTOISEBmatrixToFSLBVecs.

eurotomania commented 5 months ago

Hello,

In short, the issue is rounding errors. TORTOISE with Bmatrices in the form of : Bxx Bxy Bxz Byy Byz Bzz

The bvalue would be equal to: Bxx + Byy + Bzz

If many conversions or Bmatrix rotations occur, the sum might deviate a little bit due to rounding errors but nothing major. IT should not cause any issues. And Yes TV3.2 and T4 would have very similar behaviours. I think both represent the Bmatrix as double.

On Mon, Feb 5, 2024 at 8:57 AM ccshen @.***> wrote:

Reopened #12 https://github.com/eurotomania/TORTOISEV4/issues/12.

— Reply to this email directly, view it on GitHub https://github.com/eurotomania/TORTOISEV4/issues/12#event-11704574737, or unsubscribe https://github.com/notifications/unsubscribe-auth/AEFQZWAHACGKF5UQ7VNII63YSDQN3AVCNFSM6AAAAABC2D7FX2VHI2DSMVQWIX3LMV45UABCJFZXG5LFIV3GK3TUJZXXI2LGNFRWC5DJN5XDWMJRG4YDINJXGQ3TGNY . You are receiving this because you are subscribed to this thread.Message ID: @.***>

-- M. Okan Irfanoglu Graduate Student & GTA Computer Sciences & Engineering Ohio State University

johnccshen commented 5 months ago

Thanks for the explanation. That's close to what we observed!