[MINC-users] rawtominc & nu_correct difficulties

[Audette, Michel]

Hi Andrew, 

thanks so much for this. I don't have your experience with the so-called beast. That's definitely a good start. 

Cheers, 

Michel


-----Original Message-----
From: minc-users-bounces at bic.mni.mcgill.ca on behalf of Andrew Janke
Sent: Thu 8/2/2007 2:39 PM
To: Audette, Michel
Cc: MINC users mailing list
Subject: Re: [MINC-users] rawtominc & nu_correct difficulties
 
> looking at the data, it appears that much of the inhomogeneity is concentrated at the sides. Moreover, I only know the resolution of the image, and don't have any info about the field strength involved. Would it be appropriate to estimate the distance based on the span the locally bright area? Currently, despite converting to short, there doesn't seem to be an effect.
>
> Andrew and Vladimir, I am including an image for your perusal.
>
> All that I know about the scan is this:
>
> Human S16885
> Homo sapiens - Human
> Inner and middle ear
> Headfile available.
> 40 mm fov
> 78.125 um isotropic voxels.
> Description - Scan is at a high resolution and has great contast.
> Segmentations available
> Reconstructions available
> QuickTime Movies available

Hi again Michel,

First I would reccomend that you only correct within mask that does
not include the white border. N3 uses histogram sharpening and I
suspect that the histogram will be swamped with a very large "white"
or high value peak.

Also with the white border saturated I suspect that it would throw the
correction off anyhow given that there is no inhomogeneity information
left in there.

As for the distance, from a look at the data I would start with a
distance of about 1/2 - 1/3 of the width of the image given what
appears to be wavelength in the picture.  Of course you would be far
better of to know thie field strength! :)



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