Skip to content

New script dwicat#1626

Merged
jdtournier merged 2 commits intoMRtrix3:devfrom
jvohryzek:dwicat
Jun 18, 2019
Merged

New script dwicat#1626
jdtournier merged 2 commits intoMRtrix3:devfrom
jvohryzek:dwicat

Conversation

@jvohryzek
Copy link
Copy Markdown

Hey Robert,
Could you get it approved by some other developer.

Best
Lena & Jakub

Co-authored-by: Lena Dorfschmidt <ld548@cam.ac.uk>
Co-authored-by: Robert E. Smith <robert.smith@florey.edu.au>
@Lestropie Lestropie self-assigned this Jun 8, 2019
@Lestropie Lestropie marked this pull request as ready for review June 8, 2019 16:24
Enabled by changes in MRtrix3#1628

Provide all b=0 volumes of each input series to mrhistmatch, rather than explicitly calculating the mean b=0 image. This provides a greater amount of data for the least squares determination of scaling factor between series. This was however not possible prior to MRtrix3#1628, as previously mrhistmatch worked explicitly on 3D images only.
@Lestropie Lestropie requested a review from jdtournier June 8, 2019 16:47
@Lestropie
Copy link
Copy Markdown
Member

Closes #1428.

Copy link
Copy Markdown
Member

@jdtournier jdtournier left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Looks great, thanks!

My one comment (based on a recent discussion where this functionality will be most welcome) would be to add an option to set the average b=0 intensity - useful in cases where the data might come from different scanners where the scaling might differ wildly between subjects. But it's nothing that can't be done in a subsequent step, so don't let that stop this from merging.

@jdtournier jdtournier merged commit 5cb2397 into MRtrix3:dev Jun 18, 2019
@Lestropie
Copy link
Copy Markdown
Member

Manually setting average b=0 intensity would be pretty easy. What I actually wanted to do here, but couldn't convince myself that my math was correct, was to intensity-match all protocols to all other protocols, producing a scaling factor matrix, and then scale each protocol individually in such a way as to converge their intensities toward a common target, rather than arbitrarily using the first scan as the "target" intensity.

Lestropie added a commit that referenced this pull request Aug 9, 2019
Error in resolving divergence between #1626 and #1449.
@treanus
Copy link
Copy Markdown

treanus commented Sep 24, 2019

Hi developpers,

I did a git checkout dev and tried dwicat.

dwicat part1.mif part2.mif full.mif

Get the error:

[ERROR] mrhistmatch scale 1b0.mif 0b0.mif 1rescaledb0.mif (dwicat:84)
dwicat: [ERROR] Information from failed command:
dwicat:
        mrhistmatch: [ERROR] mrhistmatch currently only works on 3D images

@jdtournier
Copy link
Copy Markdown
Member

The script relies on updates to the supporting executables, which need to be recompiled. You need to run ./build after checking out the dev branch to get it to work.

@treanus
Copy link
Copy Markdown

treanus commented Sep 24, 2019

The script relies on updates to the supporting executables, which need to be recompiled. You need to run ./build after checking out the dev branch to get it to work.

Works fine now.

@jdtournier Thx

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants