[MINC-development] MINC and ITK

[Leila Baghdadi]

Hi everyone,

I wrote the ITK minc2 support sometime after the development of minc2 
was complete mostly because I was writing a lot of code in ITK and was 
tired of having to change file formats all the time and lose the 
minc header information. I got a few posts from people who knew I was doing both ITK 
and MINC2 development so I submitted the ITKMINC2 support to the ITK 
journal so that these people could get a hold of the code and use it. 
ITKMINC2 support makes MINC2 API calls exclusively and as a result 
unable to read MINC1 format which is not used much in our lab. As Jason mentioned, we are happily using MINC2 so the thought of supporting MINC1 never crossed my mind.

If
 you check vtk for instance, I wrote my own vtkMINC2Image reader and 
writer when we needed it in our lab and David Gobbi has committed 
various classes for reading/writing  minc1 and obj formats and so on.

I agree that the time has come to have one ITK-MINC support code which can take care of both formats,

So I think the best way is for Vladimir and I to join forces for itk-minc support and figure out what is the best way to proceed.

I will update the list as soon as I have some information.

Leila

p.s thank you Sean for your interest and starting this thread
p.p.s Thank you Jason for granting me the minc core developer privilege although I have written a lot of minc2 code, some of those guys still beat me to it, they started way before I knew what imaging is!

----- Original Message -----
From: Andrew Janke <a.janke at gmail.com>
Sent: Sun, 5/15/2011 8:29pm
To: MINC development discussion & planning <minc-development at bic.mni.mcgill.ca>
Subject: Re: [MINC-development] MINC and ITK

> That's my understanding as well - we (us Toronto folks, that is), seem to be the happiest in the MINC community to discard MINC1 compatibility,

Pick me too!  I have been MINC2 only for at least the last 3 years,
the impending MINC2.2 release will finally drive the final nail in the
MINC1 coffin. The default will be to write out MINC2 files
irrespective of what the input was.  I don't plan to add a -1 flag to
all the binaries to write out MINC1 and instead will only add a -1
flag to mincconvert for the sticklers.

For those asking why, I have had requests for this from users none of
whom were from the BIC!  Perhaps I should open a separate thread about
this on minc-dev but expect to see my commits for this on github
shortly.


a
_______________________________________________
MINC-development mailing list
MINC-development at bic.mni.mcgill.ca
http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-development